User talk:Oleg Alexandrov/Archive12

Thank You! It would be very nice to become a wikipedian but I don't understand how works many things in Wikipedia. Therefore I just correct articles which is connected to mathematics. Maybe I will be ready after some months. Also I don't know how I can create my user page and give some information about me.

I am 25 years old math student from Latvia. Also I have some results and publications in math and computer science. Thank You! Raitis Ozols from Latvia. --RaitisMath 12:00, 21 September 2007 (UTC)[reply]

Don't worry, it takes a while to learn how to get around. We appreciate any contributions, be they small fixes or new articles.

To create a user page, just click on the red link above that is your signature, and type there anything, and save it.

And one suggestion. Could you please use an edit summary when you change an article, it helps us understand what you changed. Thanks, and enjoy the wiki. Oleg Alexandrov (talk) 14:21, 21 September 2007 (UTC)[reply]

Page Reference:Pythagorean triples Comment Reference: Your comment "rv, publish that in a paper first, and then cite it. See WP:NOTABILITY"

As given in your comment, I have read "WP:NOTABILITY" guidelines but I think you can understand the fact that new manuscripts can't pass the test of "Significant coverage","Reliable Source" and "Independent of the subject".

I do not know whether you have personally visited the following page and have seen the formula: http://vcpandya.googlepages.com/pythagoreantriples But formula discussed over there can be verified by anyone. Moreover I had written it in wiki since it was related to that topic & I thought it would be useful to everyone.

I am not related to any university and I am not a professional author so publishing such a small thing is not possible so there is no possibility of meeting the criteria defined in WP:NOTABILITY guidelines. However I believe that Wiki should give more importance to truth.

If you know any sources by which I can publish my manuscript to meet the criteria then please provide some guidance for the same.

Waiting for your reply. Vikram Pandya 10:54, 22 September 2007 (UTC)

I don't know how you can publish your result. Perhaps it is already published somewhere by somebody else and you could cite that publication. Otherwise, Wikipedia does not accept material that is not already published somewhere else, since Wikipedia is a secondary source, not a primary one (so not a journal for new results). Oleg Alexandrov (talk) 14:52, 22 September 2007 (UTC)[reply]

Thanks for your reply & Minor modification to my link.
Vikram Pandya 18:02, 22 September 2007 (UTC)

Wow! This is the shortest I have ever seen your talk page.

If you have a moment, and it hasn't already been done, could you move Good article candidates to Good article nominations? The discussion is on the talk page: there is general agreement about it, but it needs an admin because the target page has three edits.

Thanks. Geometry guy 18:16, 25 September 2007 (UTC)[reply]

Done. Cheers, Oleg Alexandrov (talk) 20:12, 25 September 2007 (UTC)[reply]

Thanks. Am tidying up now. Geometry guy 20:42, 25 September 2007 (UTC)[reply]

After I did the move I looked at the redirects with the aim of fixing the double redirects, but I could not find any. I see now, I did not pay enough attention. Thanks for fixing the redirects. Oleg Alexandrov (talk) 20:55, 25 September 2007 (UTC)[reply]

I was going to add a PS promising to tidy up, but you were too fast for me! The links are a bit complicated, so I looked into them carefully before requesting the move. I hope I am catching them all now, but it may still take a day or two. Geometry guy 20:58, 25 September 2007 (UTC)[reply]

I checked again, and I still can't see any double redirects (besides the ones you already fixed). Gosh, I guess I really can't see well. Oleg Alexandrov (talk) 21:06, 25 September 2007 (UTC)[reply]

Good. I think I caught them all now, but it's nice to have another pair of eyes. Geometry guy 21:11, 25 September 2007 (UTC)[reply]

Dear Mr. Alexandrov, what is the reason for making such strange edits as [1]? You say you "removed ugly space", but I do not see what is ugly about it. My edits are like art, every pushing of the keyboard having a definite purpose and style. Did you not consider that there is a valid reason for inserting that space there? My impression is that you may be a pedantic, anal-retentive mathematician. Contrary to your supposed status as a cool applied guy. Please stop!!! Do you hate women? -- Horrible Horoball 02:23, 28 September 2007 (UTC)[reply]

The style conventions call for exactly one empty line between sections, for consistency and beauty. Yeah, I like women. Oleg Alexandrov (talk) 03:05, 28 September 2007 (UTC)[reply]

Hi ! The following pictures contain accidentally false information and are redundant since the originals are available in Wikimedia Commons. Since I've linked my article now to the originals, it might be a good idea to delete these copies. Regards --Kmhkmh 06:22, 2 October 2007 (UTC)[reply]

• Image:522px-Thales_theorem_7.png
• Image:800px-Thales_theorem_6.png

Sorry for the delay. Could you point out where on Commons are the originals? Just to make sure I don't delete something that is irreplaceable. Thanks. Oleg Alexandrov (talk) 03:33, 3 October 2007 (UTC)[reply]

I found the commons original, http://commons.wikimedia.org/wiki/Image:Thales_theorem_6.png, so I deleted the above. Oleg Alexandrov (talk) 03:53, 4 October 2007 (UTC)[reply]

Please keep an eye on Dattorro (talk · contribs) recent contributions that defy the rules of wikipedia and logic of mathematical exposition. Since they do not respond in their talk page, I blocked them until a dialog is started. Míkka 07:00, 3 October 2007 (UTC)[reply]

The stuff he's ben adding at Euclidean metric is probably a bit odd. However, looking at the history of Euclidean metric it appears to me that you misused the admin tools in blocking an editor you were in conflict with after only minor attempts at discussion on his talk page. That guy is stubborn for sure, but he's rather inexperienced as an editor, and he's no vandal, he's a prof at Stanford. Perhaps next time you could be more gentle to new editors and try to put a note at WT:WPM in such cases. Oleg Alexandrov (talk) 15:03, 3 October 2007 (UTC)[reply]

As I wrote above, I blocked him for a short period exactly in order to bring his attention to his talk page. But he ignored the request, so I unblocked him in about an hour. Usually it worked. Anyway, I was wrong. I have to refresh my knowledge of the policies more often. I am blocking "regular" editors extremely rarely. `'Míkka 00:20, 11 October 2007 (UTC)[reply]

...he's a prof at Stanford. Actually, no. He graduated with a Ph.D. in EE in 2005. I assume he has some kind of research associate position, or at least, they seem to be letting him keep his office space and phone #. I only mention this as your idea about his status seems to be affecting your views of this situation, so you should at least know the actual status of Dattoro. --Horoball 21:41, 12 October 2007 (UTC)[reply]

Please see Wikipedia talk:Requests for adminship#Subheadings for discussion sections for a centralized discussion about adding some additional sub-headings to {{RfA}}.

I'm alerting you because you're listed at the top of Template talk:RfA as someone that needs to be informed of any changes; if this change will break your bot, I'd like to know before doing it. :) EVula // talk // ☯ // 18:07, 11 October 2007 (UTC)[reply]

I replied there. Oleg Alexandrov (talk) 03:51, 12 October 2007 (UTC)[reply]

I don't understand why one would use American spelling when the version of Wikipedia is clearly stated as English and not American. English is the language of England. It makes as much sense to put in some other invented spelling. It's very confusing for non-native speakers when reading articles. —Preceding unsigned comment added by 140.203.8.86 (talk) 16:37, 15 October 2007 (UTC)[reply]

What Titoxd wrote on your talk page. Oleg Alexandrov (talk) 03:38, 16 October 2007 (UTC)[reply]

If I remove a link from one of these lists (without turning it blue), will mathbot put it back later? I'm thinking in particular of Additive inverse of an inverse element, but I'd say that many planetmath articles just don't have appropriate titles, and probably don't even need to be redirects, since no one is likely to search for them. But I don't want to take the time to remove them from the list if they'll just be put back. --Sopoforic 10:26, 18 October 2007 (UTC)[reply]

The bot would normally put it back. To prevent that from happening, you can add it to Wikipedia:Missing science topics/Blacklisted. You are right that there exist many poor titles out there. Oleg Alexandrov (talk) 15:21, 18 October 2007 (UTC)[reply]

Hi. I'd like to request a change be made to the formatting of the statistics generated by the bot, matching this change. This will remove some excessive whitespace after the table when it is transcluded (and remove some redundant bolding). Thanks in advance. --PEJL 19:31, 19 October 2007 (UTC)[reply]

Continued at WT:1.0/I. Oleg Alexandrov (talk) 03:31, 20 October 2007 (UTC)[reply]

I noticed that the lists have a bug - whenever an article title has an ampersand in it, the lists include the title twice. One time the title has a real ampersand, the second time it has & instead. The affected titles are:

• Journal of Business & Economic Statistics
• Journal of Computational & Graphical Statistics
• National Assessment & Testing
• ZAMM — Journal of Applied Mathematics & Mechanics

— Carl (CBM · talk) 13:22, 20 October 2007 (UTC)[reply]

Thanks for pointing this out. At some point query.php started using & instead of plain &, and I guess that's how things got duplicated. I made a fix to the bot now to that respect, let's see what happens tomorrow. Oleg Alexandrov (talk) 03:43, 21 October 2007 (UTC)[reply]

I also had to fix bugs in my code that were caused by that change. I don't know the reason, but I saw claims that the extra HTML escapes were added because of a security risk in the original code. The same changes were applied to api.php. — Carl (CBM · talk) 04:10, 21 October 2007 (UTC)[reply]

Hi - the WP 1.0 bot appears no longer to work on Category:Good articles by quality: the last update was on 14 October. I don't think there have been any changes at GA. Have you modified the script recently? It is possible that the fact that the category is not fully populated by the standard subcategories is now causing problems. If that's all it is, I can recreate the subcategories that were deleted after the silly AfD this summer. Geometry guy 20:03, 22 October 2007 (UTC)[reply]

PS. Also, GA doesn't need importance categories. Can these be deleted? PPS. I only just noticed you have a new job. I hope it is working out well.

I'll look into this, hopefully tomorrow. Oleg Alexandrov (talk) 03:29, 23 October 2007 (UTC)[reply]

That was a mistake I did, and it affected all the projects. I fixed it now. Thanks for pointing it out. Oleg Alexandrov (talk) 02:56, 25 October 2007 (UTC)[reply]

Have you started coding FL and list class?--TonyTheTiger (t/c/bio/tcfkaWCDbwincowtchatlotpsoplrttaDCLaM) 17:26, 23 October 2007 (UTC)[reply]

There was no agreement on this, as far as I could see. Some people said that we don't need these to start with, others wanted a whole lot of other classes in addition. I don't care much myself, but I'd rather not code something there is no agreement on. Oleg Alexandrov (talk) 02:17, 24 October 2007 (UTC)[reply]

See this thread for more info. Many thanks! Girolamo Savonarola 23:25, 24 October 2007 (UTC)[reply]

I replied there. Oleg Alexandrov (talk) 02:57, 25 October 2007 (UTC)[reply]

Hello. I wonder if you'd help with/consider/give an opinion about a request for work to be done by a bot at Wikipedia:Bot requests#RfA edit counts/analysis. Since RfA candidate's edit counts are on just about all RfA talk pages and Mathbot already does valuable RfA work, could Mathbot gather edit counts and analyses and post them on the RfA talk pages automatically? WODUP ^(?) 02:13, 29 October 2007 (UTC)[reply]

I'll reply there (in short, it should be easy enough to add this functionality to the bot). Oleg Alexandrov (talk) 04:33, 29 October 2007 (UTC)[reply]

I see your bot covered the 2005 (Jan. 2006) and 2006 ArbCom elections. We are currently setting up the 2007 elections. Could your bot perform the same tasks for this new election? - Mtmelendez ^(Talk) 02:20, 29 October 2007 (UTC)[reply]

Sure, I'll keep an eye on the preparations and modify the bot code, if necessary, from what was used last year. If you could remind me again when we are getting close to the election, that would help too (but I'll try not to forget myself). Oleg Alexandrov (talk) 04:29, 29 October 2007 (UTC)[reply]

No problem. I'm gonna try to oversee the election process, making sure everything goes smoothly with the sub-pages, transclusions, and templates. I'll let you know about the last week of November, which is the last week of the candidacies. The elections begin on December 3. Thanks for your help! - Mtmelendez ^(Talk) 02:08, 31 October 2007 (UTC)[reply]

We're five days away from the voting phase. Just a quick reminder! :) - Mtmelendez ^(Talk) 01:40, 28 November 2007 (UTC)[reply]

I set up the bot to run every half an hour from now (so that I don't forget later, or if I am not online when the election starts). It outputs the results at User:Mathbot/ArbCom Election December 2007. The bot assumes that the candidate pages will be at Wikipedia:Arbitration Committee Elections December 2007/Vote/Candidate name, as last year. If you plan to use something different, please let me know so that I can adjust the bot code. Cheers, Oleg Alexandrov (talk) 07:16, 28 November 2007 (UTC)[reply]

Great! Once the election starts, I'll monitor its results for the first day or two. I'll let you know, but I don't expect major problems. Thanks!! - Mtmelendez ^(Talk) 23:46, 1 December 2007 (UTC)[reply]

FYI, it looks like the bot is looking at red link pages and producing a null result output. NoSeptember 00:45, 3 December 2007 (UTC)

That's because the election did not start yet. I'll be busy in the next several days, so I started the bot in advance. Oleg Alexandrov (talk) 00:59, 3 December 2007 (UTC)[reply]

Voting started an hour ago. No worries, the other bot has it covered for now. NoSeptember 01:08, 3 December 2007 (UTC)

Jim Belk, a decent mathematician new to Wikipedia, found himself over and over and over trying to satisfy Paolo, who again and again and again did not understand the mathematics of functions nor the mathematics of inverse functions nor the English being used to explain either one. He kept trying to "help" by rewriting both the function and the inverse function article, each time mangling both the language and the mathematics. Jim finally had enough. I came up with this wording to put an end to the mess, and it seems to have done so. There are several different meanings for inverse function, including left-, right-, two-sided; this opening covers them all without getting deep into technicalities. Language about "reverses" caused horrible problems was rejected for good reason. I haven't the patience to cover this ground again. It's good enough, and I'll guarantee you an attempt to "improve" will backfire and lead to something worse. Please do not reopen this can of worms. --KSmrq^T 08:36, 29 October 2007 (UTC)[reply]

Thanks for Changing Misc. to Miscellaneous in Eric E Schmidt, cause I'm too lazy to spell :D Compwhizii 00:18, 31 October 2007 (UTC)[reply]

Is there a way to set perlwikipedia object to send data encoded as utf8 with appropriate header? I did not found Unicode::Utf8simple or obviously Win32::MBCS on the toolserver. Mashiah 19:30, 31 October 2007 (UTC)[reply]

Sorry, I've found a way. Mashiah 23:35, 31 October 2007 (UTC)[reply]

Cool. :) I actually wouldn't know how to answer this question. Oleg Alexandrov (talk) 14:32, 1 November 2007 (UTC)[reply]

The only packet needed is Encode. I just decode('utf8',$text) and it works for me. Now fighting with non-latin characters in login name. Will report on success. Mashiah 14:36, 1 November 2007 (UTC)[reply]

You might be able to go a long with with commands like

binmode STDOUT, ":utf8";

which tell perl that all output sent to STDOUT should be utf8. You can set the binmode of any filehandle, input or output. Often just setting the modes of filehandles will allow you to ignore encodings in the rest of your script (at least on unix). — Carl (CBM · talk) 14:47, 1 November 2007 (UTC)[reply]

I think this should work too. But for now I need to know how to login with username like 'Голем', which is cyrillic. Everything works with latin names but with cyr bot makes edits with toolservers IP. Mashiah 15:31, 1 November 2007 (UTC)[reply]

Also done. It needs the username typed in utf8, unlike for urls, which are allowed as n a user locale (1251 for me). Mashiah 20:31, 1 November 2007 (UTC)[reply]

All previous were for 0.9. In 1.0 all the urls changed the encoding. Be aware =). Mashiah 00:16, 2 November 2007 (UTC)[reply]

Thanks for the info; I don't use perlwikipedia much. — Carl (CBM · talk) 03:43, 2 November 2007 (UTC)[reply]

I think the bot may be malfunctioning, as it has made over 50 edits to this page within a five minute span, alternatively adding logs for November 1 and October 31. -- Scorpion⁰⁴²² 01:31, 2 November 2007 (UTC)[reply]

I have blocked the bot. Please see here. Cheers TigerShark 01:55, 2 November 2007 (UTC)[reply]

The reason for the confusion is because two versions of the bot were running in parallel. I killed one of them, and will now unblock the bot (I'll keep an eye on it for a while to see how it is doing). I'll make sure this issue won't show up again. Oleg Alexandrov (talk) 02:52, 2 November 2007 (UTC)[reply]

And thanks for noticing and the prompt action. Oleg Alexandrov (talk) 02:54, 2 November 2007 (UTC)[reply]

Zscout pointed out to me that the WP 1.0 table shows 1933 featured articles but the FA page shows only 1665. Do you happen to have a script to cross-check that? — Carl (CBM · talk) 23:07, 2 November 2007 (UTC)[reply]

I have no quick script, although the bot source code (see the link at bot's page) can be modified to do that. Is it something of high priority or just curiosity? I hope to get to this at some point once I do the current request for adding List-Class to the bot. Oleg Alexandrov (talk) 03:00, 3 November 2007 (UTC)[reply]

Oleg, all these complicated processes are for potentially controversial things or related to massive changes. I would never dare to move something where heated discussions are abundant or expected. Here, in technical areas, where there is no drama I am bold quite often, especially in areas of my profession. Right now I am writing the missing microfabrication article. In fact, I am a very reasonable man, unless someone's arrogance really pisses me off, and if you had told me to back off (with a reasonable argument) here, I'd have undone/redone my changes. `'Míkka 04:16, 3 November 2007 (UTC)[reply]

Yeah. :) Oleg Alexandrov (talk) 14:36, 3 November 2007 (UTC)[reply]

It seems the bot isn't updating the quality rowspan value in the table, as shown here. All in all though, kudos with the bot, it'd be impossible to evaluate quality without it. --TMF ^{Let's Go Mets - Stats} 08:28, 5 November 2007 (UTC)[reply]

I am not sure I understand what you are trying to say. Can you point out to a specific value in a cell which should be something else and is not? The table looks OK to me so far. Oleg Alexandrov (talk) 15:44, 5 November 2007 (UTC)[reply]

The entire bottom row is off by one cell... --TMF ^{Let's Go Mets - Stats} 16:07, 5 November 2007 (UTC)[reply]

Basically, the quality cell has "rowspan=10" when it should be "rowspan=11". The bot added a row for list-class (honestly, I'm not sure if this is a worthy addition, especially because it assumes that a List-Class article has an importance rating, but I digress) without increasing the rowspan of the quality cell. --TMF ^{Let's Go Mets - Stats} 16:27, 5 November 2007 (UTC)[reply]

Looks like it fixed itself - [2] --TMF ^{Let's Go Mets - Stats} 16:49, 5 November 2007 (UTC)[reply]

Fix itself ..., that's too much to hope for. :) Thanks for pointing this out! Oleg Alexandrov (talk) 16:56, 5 November 2007 (UTC)[reply]

Dear Oleg, thanks for your support. I do not have a photomask picture (yet), but if I do get one, I will still need to ask for permission to post. Will be looking...Sincerely, Guiding light 04:25, 6 November 2007 (UTC)[reply]

Replied on your talk. Oleg Alexandrov (talk) 04:56, 6 November 2007 (UTC)[reply]

Thanks so much for your picture edit! It looks much better! I was thinking about re-submitting the file...thanks again..Guiding light 05:00, 6 November 2007 (UTC)[reply]

Is there any way that Mathbot could add the rest of the edit count details? The namespace breakdown is great, but the most edited article information is also an extremely helpful launching pad for investigating an editor's contributions. EVula // talk // ☯ // 17:45, 6 November 2007 (UTC)[reply]

Sure. I'll implement this in a few days. Oleg Alexandrov (talk) 06:12, 7 November 2007 (UTC)[reply]

Excellent, thank you. :) EVula // talk // ☯ // 01:42, 8 November 2007 (UTC)[reply]

(Much belated) I implemented this, as seen here. It works as a copy and paste from Interiot's tool (so all the HTML markup is removed). Oleg Alexandrov (talk) 06:09, 13 November 2007 (UTC)[reply]

Beautiful, it's everything I could hope for. :) EVula // talk // ☯ // 06:15, 13 November 2007 (UTC)[reply]

See the history of User:Mathbot/Most recent admin candidate. It seems to be posting information about User:Spellcast's RfA every 15 minutes for some reason. --ais523 18:49, 7 November 2007 (UTC)

That was due to the rather rare event that both the nomination page and its talk page have deleted versions in the history (the routines I use for uploading can't handle that). I restored the deleted versions and then the bot did his job and stopped trying. Thanks for pointing this out! Oleg Alexandrov (talk) 04:32, 8 November 2007 (UTC)[reply]

It happened again: Wikipedia:Requests for adminship/Oxymoron83. I've restored the deleted revision on the Talk page in the hope that that will stop the problem in this case. --ais523 13:31, 12 November 2007 (UTC)

I modified the bot to add the text to Wikipedia talk:Requests for adminship if both the Rfa candiate page and its talk page has problems as above. In the worst case, it will post to my talk page. Hopefully that will fix the problem for good. Thank you for your report! Oleg Alexandrov (talk) 04:42, 13 November 2007 (UTC)[reply]

Hi Oleg, здраствуй,

I am not too concerned about epsilons :-). In general, the wikipedia should be more careful about "necessary". Mathematics is at its best partly when it arrives at short and widely recognized notation. Thus I wish that some of the most common mathematical notation would be accepted by wikipedia just as standard English is. This would help to avoid uglifying the mathematics. Best regards, Wlod 01:57, 8 November 2007 (UTC)[reply]

But please note that Wikipedia is not a scientific paper or a math book. It is a general purpose encyclopedia. As such, complicated notation should not be employed unless really necessary. I do agree however that given a professional math paper you want to set the notation early. Oleg Alexandrov (talk) 04:34, 8 November 2007 (UTC)[reply]

The goal of notation is to make things simpler and more readable. Avoiding the notation makes the articles less readable.

Some mathematical notation should be a part of every educated person. I would include as a part of a standard everybody's language such mathematical phrases as  ${\displaystyle a\in A\subseteq B}$. Replacing this phrase by words buys nothing – those who cannot absorb this phrase cannot get anything even from mildly advanced mathematical articles.

Also, wikipedia is for everybody, but not all articles should be meant for everybody, it's totally unrealistic. For instance, the statement about the distribution of the prime numbers is meant only for a relatively small fraction of the wikipedia audience. Thus there is no reason to obscure the article, and thus to make it less valuable (due to its messy and clumsy look) by refraining from the mathematical notation.

Some historical and biographical articles may avoid the technical notation. But technical articles should employ it. -- Wlod 08:22, 8 November 2007 (UTC)[reply]

Oleg, if you have time, would you be so nice and have a look at the article. Any comments would be welcome. I've written that article because there were some in wikipedia already about the related topics about metric spaces, including the injective metric spaces (i.e. hyperconvex), which have applications to Banach spaces. I didn't give the full justice to the topic since it is but an encyclopedic note. On the other hand, when wikipedia mentions the related topics then it would be incomplete in an essential way by ommiting the given topic, especially when it has a direct, geometric appeal. Best regards, Wlod 01:57, 8 November 2007 (UTC)[reply]

I am not sure the article has a purpose. The article should be focused on a specific topic, appropriately named after that topic, and then describing it. What you wrote looks like some kind of collection of thoughts on a few things rather than an article about something specific. I'd like to note that the encyclopedic style has certain restrictions, and I am not sure how appropriate your article is the way it is written. Oleg Alexandrov (talk) 04:36, 8 November 2007 (UTC)[reply]

Did you actually follow the mathematics of the article? (It's easy but it still has to be done to form an opinion about the article). Yes, the purpose of the article is very clear: among all superspaces aiming at the given arbitrary metric space, there is a maximal one). -- Wlod 08:30, 8 November 2007 (UTC)[reply]

Thanks for your tips on using internal links etc! Wschaap —Preceding unsigned comment added by Wschaap (talk • contribs) 09:32, 9 November 2007 (UTC)[reply]

Hi Oleg, I deleted the section November 7, 2007 from Wikipedia:Version 1.0 Editorial Team/Pink Floyd articles by quality log (Version Log: 05:16, 7 November WP 1.0 bot (Talk | contribs) (6,047 bytes) (Log for November 7, 2007) – http://en.wikipedia.org/w/index.php?title=Wikipedia:Version_1.0_Editorial_Team/Pink_Floyd_articles_by_quality_log&oldid=169791878 – on 7 November, because of a bot error (or my error, but I didn't find one). The bot added 14 articles related to the Clash in that log page, that is part of the Pink Floyd WikiProject.

Deleted section

===November 7, 2007===
* [[(White Man) In Hammersmith Palais]] ([[Talk:(White Man) In Hammersmith Palais|talk]]) Start-Class (No-Class) added.
* [[Capital Radio One]] ([[Talk:Capital Radio One|talk]]) Start-Class (No-Class) added.
* [[Concerts for the People of Kampuchea (album and EP)]] ([[Talk:Concerts for the People of Kampuchea (album and EP)|talk]]) Start-Class (No-Class) added.
* [[From Here to Eternity: Live]] ([[Talk:From Here to Eternity: Live|talk]]) Start-Class (No-Class) added.
* [[Groovy Times]] ([[Talk:Groovy Times|talk]]) Start-Class (No-Class) added.
* [[Jail Guitar Doors]] ([[Talk:Jail Guitar Doors|talk]]) Start-Class (No-Class) added.
* [[Live at Bond's Casino]] ([[Talk:Live at Bond's Casino|talk]]) Start-Class (No-Class) added.
* [[Mick Jones (The Clash)]] ([[Talk:Mick Jones (The Clash)|talk]]) Start-Class (No-Class) added.
* [[Paul Simonon]] ([[Talk:Paul Simonon|talk]]) Start-Class (No-Class) added.
* [[Rude Boy (film)]] ([[Talk:Rude Boy (film)|talk]]) Start-Class (No-Class) added.
* [[The Cost of Living (EP)]] ([[Talk:The Cost of Living (EP)|talk]]) Start-Class (No-Class) added.
* [[The Essential Clash]] ([[Talk:The Essential Clash|talk]]) Start-Class (No-Class) added.
* [[The Singles (2007 album)]] ([[Talk:The Singles (2007 album)|talk]]) Start-Class (No-Class) added.
* [[Topper Headon]] ([[Talk:Topper Headon|talk]]) Start-Class (No-Class) added.

Today, the bot wrote a new (and correct) log (Version Log: 05:57, 9 November 2007 1.0 bot (Talk | contribs) (6,159 bytes) (Log for November 9, 2007) with the "removed" message for all of the previous 14 pages, but the 7 November section isn't visible on that page (because of my deletion).

Section added by the bot

===November 9, 2007===
* '''[[(White Man) In Hammersmith Palais]] ([[Talk:(White Man) In Hammersmith Palais|talk]]) Start-Class (No-Class) removed.'''
* '''[[Capital Radio One]] ([[Talk:Capital Radio One|talk]]) Start-Class (No-Class) removed.'''
* '''[[Concerts for the People of Kampuchea (album and EP)]] ([[Talk:Concerts for the People of Kampuchea (album and EP)|talk]]) Start-Class (No-Class) removed.'''
* '''[[From Here to Eternity: Live]] ([[Talk:From Here to Eternity: Live|talk]]) Start-Class (No-Class) removed.'''
* '''[[Groovy Times]] ([[Talk:Groovy Times|talk]]) Start-Class (No-Class) removed.'''
* '''[[Jail Guitar Doors]] ([[Talk:Jail Guitar Doors|talk]]) Start-Class (No-Class) removed.'''
* '''[[Live at Bond's Casino]] ([[Talk:Live at Bond's Casino|talk]]) Start-Class (No-Class) removed.'''
* '''[[Mick Jones (The Clash)]] ([[Talk:Mick Jones (The Clash)|talk]]) Start-Class (No-Class) removed.'''
* '''[[Paul Simonon]] ([[Talk:Paul Simonon|talk]]) Start-Class (No-Class) removed.'''
* '''[[Rude Boy (film)]] ([[Talk:Rude Boy (film)|talk]]) Start-Class (No-Class) removed.'''
* '''[[The Cost of Living (EP)]] ([[Talk:The Cost of Living (EP)|talk]]) Start-Class (No-Class) removed.'''
* '''[[The Essential Clash]] ([[Talk:The Essential Clash|talk]]) Start-Class (No-Class) removed.'''
* '''[[The Singles (2007 album)]] ([[Talk:The Singles (2007 album)|talk]]) Start-Class (No-Class) removed.'''
* '''[[Topper Headon]] ([[Talk:Topper Headon|talk]]) Start-Class (No-Class) removed.'''

So, or the section for 7 November must be included newly or the section created by the bot for 9 November may be deleted, or I don't know??? I am sorry, and have a nice weekend — Pjoef 19:19, 9 November 2007 (UTC)[reply]

Uhm, I don't know. Let's see if there are any issues in the future. Oleg Alexandrov (talk) 10:34, 16 November 2007 (UTC)[reply]

Lists of basic topics, Lists of topics, and List of glossaries were moved to portal space without any prior discussion. A discussion is underway at Wikipedia talk:Lists (see the topic on the contradiction with ASR); the mathematics lists are also being discussed there. I should have moved those pages back directly but corrected the portal titles instead. An admin is needed to return those lists to their locations in the main namespace. They should be in the main namespace where they were until a consensus is reached to put them somewhere else. Would you move them back please? The Transhumanist 22:23, 9 November 2007 (UTC)[reply]

Has the bot been running more quickly than usual? --Rschen7754 (T C) 07:41, 16 November 2007 (UTC)[reply]

I see something like 4 edits per minute, which is just about right. Or do you think that's too fast? Oleg Alexandrov (talk) 10:36, 16 November 2007 (UTC)[reply]

I've just noticed that it usually took 3-5 days for the USRD statistics to get updated. Now it's updating daily. --Rschen7754 (T C) 23:03, 16 November 2007 (UTC)[reply]

Could you provide a link to what you mean by the USRD statistics? My best guess is that somebody is running the bot by hand from the web-based user interface (as mentioned in the instructions). Oleg Alexandrov (talk) 00:36, 17 November 2007 (UTC)[reply]

Hello, I am Happy5214!! I am writing to tell you that WP 1.0 bot is malfunctioning!! The error message is "Could not get_text for Wikipedia:WikiProject New Hampshire State Highways]]." Please help!!

-Happy5214 (talk) 23:15, 16 November 2007 (UTC)[reply]

It is not bot's fault per se, it was confused by a syntax error which I fixed. The bot works now. Thanks for pointing this out! Oleg Alexandrov (talk) 23:41, 16 November 2007 (UTC)[reply]

When Mathbot updates Wikipedia:Articles for deletion/Old is it possible for the edit summary to indicate if theres a backlog. currently it just says Statistics of open discussions no matter how many there are. Gnangarra 04:32, 22 November 2007 (UTC)[reply]

Done, nice idea! Now the bot states there is a backlog if the number of open discussions is over 200. Is that too high? Oleg Alexandrov (talk) 05:09, 22 November 2007 (UTC)[reply]

thanks, let it run with 200 for a while i think it might to low rather than too high the daily afd nominations run between 150 -180 anyway. Gnangarra 06:22, 22 November 2007 (UTC)[reply]

Dear Oleg,

I have come across you paper on the Green's function of a circular dielectric fiber. it containes a reference to my old paper on the radiation condition for open waveguides. Some time later I published another, more elaborated paper on this subject, however in a less known journal, "J. of EM Waves and Aplications". If you are still interested in this topic, I can send you a copy. So, if you like, write me to <snipped>. Currently I ma on a visit in Nottingham, going back to Ukraine tomorrow morning.

Sincerely, Alexander Nosich www.ire.kharkov.ua/dep12/MOCA —Preceding unsigned comment added by 128.243.220.41 (talk) 11:18, 22 November 2007 (UTC)[reply]

Will reply by email. Oleg Alexandrov (talk) 11:54, 22 November 2007 (UTC)[reply]

Hi. WP 1.0 bot keeps removing {{Unassessed-Class|category=Category:Unassessed Hawaii articles|Unassessed}} from Wikipedia:Version 1.0 Editorial Team/Hawaii articles by quality statistics [3] While I thought this might be normal at first, since I've been trying to stay on top of the category and keep it empty, I've noticed that even when the category is populated, the bot doesn't add it back in. Any idea what's going on here? I've reverted for a second time,[4][5] but obviously the bot will remove it on the next update. Thanks for your help. —Viriditas | Talk 07:28, 24 November 2007 (UTC)[reply]

Hi. The bot should not remove this row in the table if it is populated. Try populating this category and running the bot by hand, and see what happens. The bot removes empty rows per some discussion we had recently at WT:1.0/I. Oleg Alexandrov (talk) 18:24, 24 November 2007 (UTC)[reply]

You are right, of course. I ran the test and it worked great. My error was in failing to recognize the bot's 72 hour schedule, whereas I was checking the category anywhere between 24-48 hours. Thanks for clearing this up for me. —Viriditas | Talk 04:38, 25 November 2007 (UTC)[reply]

It looks like the problem has returned again. Based on this, it looks like the bot isn't counting the number of classes correctly when it compiles the table. For some reason, the bot insists on having the quality cell span 11 rows instead of 8 (which would be correct in this case). I've also seen this problem on a lot of projects that have no featured articles or A-Class articles (and thus no rows for them). --TMF ^{Let's Go Mets - Stats} 22:41, 25 November 2007 (UTC)[reply]

The problem is different now though, before the number of rows were too few, now they are too many (this latter thing does not affect things much though I think). I now modified the code to actually calculate the number of rows rather than relying on an upper bound. Thanks for the note. Oleg Alexandrov (talk) 00:48, 26 November 2007 (UTC)[reply]

I have been dabbling with {{WikiProject Southern California}}. My main aim was to change the template so that it would list any articles tagged for both WikiProject Southern California and WikiProject California. I have had some success, but the work is incomplete. It seems that the template DOES jointly list quality, but does not list importance for both. See Talk:Balboa Pavilion for an example. Can you recommend a porgrammer I can work with to fix this? If I can get it to work, I'm going to convert the template for {{WikiProject Philadelphia}} as well. --evrik ^(talk) 16:35, 26 November 2007 (UTC)[reply]

Uhm, I don't know. Try to see who wrote the Military History and/or Biography templates, those were tricky I think, and whoever did them could help. Or ask at WT:1.0/I, that page is watched by many folks who know these things. Oleg Alexandrov (talk) 03:44, 27 November 2007 (UTC)[reply]

Oleg, Your advice welcomed regarding an update to the page on Solid Angle. I wish to add some extra formulae at the end of the paragraph relating to Solid angle in arbitrary dimension. I have prepared the update in my Sandbox - user:fjackson/sandbox. The reason for the update is that the Gamma Function is difficult to use. The Formulae I have added are more helpful in calculating the higher dimension values. The only reference that I have is my own work on the subject and I am conscious of "signing" the article. Regards Frank M Jackson (talk) —Preceding comment was added at 19:22, 26 November 2007 (UTC)[reply]

Good you asked. :) The policies on Wikipedia don't encourage original research. However, looking at User:Fjackson/Sandbox and the manner you asked about it, I think it is OK if you add that text. I put solid angle on my watchlist, and will help with copyediting or if other people try to challenge the addition. Cheers, Oleg Alexandrov (talk) 03:47, 27 November 2007 (UTC)[reply]

Could you tell me, why you are thinking, that ${\displaystyle X(\omega )={\begin{cases}0.5,&\omega ={\texttt {H}},\\0.5,&\omega ={\texttt {T}}.\end{cases}}}$ is a good example for a random variable. This may be the probability for the coin-toss but not a random variable. It should be 1,0 or anything that differs. Am I wrong ? —Preceding unsigned comment added by 129.206.107.233 (talk) 15:11, 28 November 2007 (UTC)[reply]

OK, I think you are right. How about 0.7, 0.3 then? I'll put that in the article. Oleg Alexandrov (talk) 15:44, 28 November 2007 (UTC)[reply]

Perhaps there is some confusion between a random variable and a probability distribution? Another example for a random vairable (a bit more conventional) would be

${\displaystyle X(\omega )={\begin{cases}1,&\omega ={\texttt {H}},\\-1,&\omega ={\texttt {T}}.\end{cases}}}$

-- Meni Rosenfeld (talk) 08:50, 30 November 2007 (UTC)[reply]

Looks random enough to me. :) Oleg Alexandrov (talk) 15:56, 30 November 2007 (UTC)[reply]

I'm with Meni. 0.7 and 0.3 is again to easy to be confused with the probability (as it sums up to 1). So -1,1 is the much better example. 1(==H) and 0(==T) have the advantage that one can easily calculate the probabilities for Head (or better spoken its approximation) from an coin toss experiment-series, by just summing up the random variable and diving it by the number of single experiments . Martin —Preceding unsigned comment added by 129.206.107.233 (talk) 13:30, 6 December 2007 (UTC)[reply]

Good points. I modified the variable to

${\displaystyle X(\omega )={\begin{cases}3,&\omega ={\texttt {H}},\\-5,&\omega ={\texttt {T}}.\end{cases}}}$

Feel free to change it yourself to anything else. Oleg Alexandrov (talk) 04:39, 7 December 2007 (UTC)[reply]

I made a couple changes to Limit superior and limit inferior and made some comments on its talk page. I'd like to hear what you have to say. Dfeuer (talk) 03:14, 29 November 2007 (UTC)[reply]

I replied there. Basically, I am not an expert in the more sophisticated aspects of lim sup. As long as the first part of the article is kept elementary, I am fine with anything else you may do afterwards. Oleg Alexandrov (talk) 04:20, 29 November 2007 (UTC)[reply]

you spelt marilyn monroes name wrong —Preceding unsigned comment added by 89.242.80.13 (talk) 15:12, 2 December 2007 (UTC)[reply]

I don't think that was me. I did not edit anything related to her for a very long time. Oleg Alexandrov (talk) 22:56, 2 December 2007 (UTC)[reply]

There seems to be a problem with this page, as the vote pages appear to be all wrong... TML 02:34, 3 December 2007 (UTC)[reply]

Thanks for pointing it out. I fixed it now. Oleg Alexandrov (talk) 04:11, 3 December 2007 (UTC)[reply]

See Wikipedia:Articles for deletion/Error-correcting codes with feedback.

Normally I would expect this to get listed at Wikipedia:WikiProject Mathematics/Current activity. But it hasn't. It bears the "Information theory" category. Apparently articles with that category don't show up on the current activity page. I've added the "Mathematics" category for now.

Oleg, are you the person who manages the software that determines which things show up on the "current activity" page? Can you do something about this? Michael Hardy 03:19, 3 December 2007 (UTC)[reply]

I am not sure if the articles in Category:Information theory can be considered math. Let's ask at WT:WPM. Oleg Alexandrov (talk) 06:18, 3 December 2007 (UTC)[reply]

This raises a related question: Is there any way for us to know WHICH categories are included among those that show up on the "current activity" page? If I want to suggest the inclusion of one, how would I know whether it is already included or not? Michael Hardy 16:27, 3 December 2007 (UTC)[reply]

The list is at List of mathematics categories. Oleg Alexandrov (talk) 03:27, 4 December 2007 (UTC)[reply]

Hi Oleg,

Thanks for discussing the constant article's renaming. I have been working on the article a bit recently and it may be a good time to ask experienced people what they think about it, so that I wouldn't be going in too bad a direction. :) Randomblue 16:40, 4 December 2007 (UTC)[reply]

I suggest you ask at WT:WPM again. Usually people give insightful comments there. Cheers, Oleg Alexandrov (talk) 06:49, 5 December 2007 (UTC)[reply]

Mr. Oleg,

Why are you removing my contributions to standard deviation?

I am writing out a step by step procedure which may help the novice. It is a lot better than the technical jargon mentioned there. —Preceding unsigned comment added by 70.108.154.228 (talk) 17:35, 4 December 2007 (UTC)[reply]

Please register as a user so that you have a stable named account. That way we can easily see which edits are yours, and you can build a reputation (hopefully for good work). Thank you. JRSpriggs (talk) 00:49, 5 December 2007 (UTC)[reply]

I removed only once. Second time I did some copyediting on it. I did not like the way you wrote it, but overall, I do agree that that such material helps. Oleg Alexandrov (talk) 06:45, 5 December 2007 (UTC)[reply]

Hi Oleg, I noticed that you voted against my arbitration candidacy on the basis that I had voted against others in the election, and I wanted to respond briefly to that. I chose to vote in this election, as I have every other year, because I care about who gets elected - whether that be who I serve alongside if I get elected, or who winds up doing the job if I don't. As such, I've supported nearly all the candidates with any chance of being elected - and I initially supported the one that I subsequently decided to vote against. If you have an objection to candidates voting at all, then I understand, but I'm concerned that a few people seem to have assumed that I must have voted as some sort of campaign tactic, when I'm actually supporting all of the people I stand to potentially lose to. Rebecca (talk) 06:03, 5 December 2007 (UTC)[reply]

No, I did not believe it was a campain tactic on your part. I objected because you chose to vote in an election you also participate in. No biggie, indeed, but I would have expected an arbitrator-to-be not to vote to avoiod any appearance of improper behavior. Oleg Alexandrov (talk) 06:58, 5 December 2007 (UTC)[reply]

Hi Oleg, and sorry if you have heard this before, but: I'd really appreciate it if you could Mathbot provide links to all open discussions on WP:AFDO, even if there are more than 20 or so of them per page. That's because:

• Wikipedia is not paper: we have enough space for all of these links. Providing them in the form of a sortable table would be a possibility, too.
• Loading huge AfD pages takes several seconds even on my broadband connection, and finding 20 open AfDs among 150 or so takes even longer. This can be avoided with the WP:AFDO shortlinks - popups even give a tooltip info what the deletion discussion is about.

Thanks for your consideration, Sandstein (talk) 07:46, 9 December 2007 (UTC)[reply]

I did it once, and then I received comments that that list is too long. Also note that WP:AfD/Old is also transculded from WP:AfD, and there a long list is not necessary. Perhaps you could start a discussion on the issue. My own view is that using some javascript popups for individual users would be a better solution. But again, you can start a discussion on this and see what people say. Oleg Alexandrov (talk) 21:10, 9 December 2007 (UTC)[reply]

I agree that this feature would be very helpful. It takes me sometimes more than 2 minutes to load the logpages. One possible solution would be to provide on the mathbot page where it says "done one may now return to...", to put the wikimarkup for the oldest 25 AFD's in text format, so a user could copy and paste it to his/her own userspace page. I would find this feature extremely useful. It would be probably nice to make it as both transclusion and list format, as some users might like it one way and others the other. A side-by-side link to the article itself in the list format would help identify any that were already deleted.

Transclusion format: {{Wikipedia:Articles for deletion/oldest AFD}} {{Wikipedia:Articles for deletion/next oldest AFD}}
and so on.

List format: * [[Wikipedia:Articles for deletion/oldest AFD]] [[Oldest AFD article]] * [[Wikipedia:Articles for deletion/next oldest AFD]] [[Next oldest afd article]] and so on. JERRY ^talk _contribs 17:24, 12 January 2008 (UTC)[reply]

Ah, this is much more than what I had envisioned for mathbot's modest tool. There used to be a separate bot somebody ran which would generate tables for each day. I don't know what happened to it. I will ask at WT:AFD, perhaps we can continue there. Oleg Alexandrov (talk) 21:45, 12 January 2008 (UTC)[reply]

Thanks for the reply. I'll just make my own VB program to do this just for me. JERRY ^talk _contribs 22:35, 12 January 2008 (UTC)[reply]

Belately (and most likely you won't see this reply), I added a new feature at WP:AfD/Old linking to a complete list of open discussions. Oleg Alexandrov (talk) 06:08, 21 January 2008 (UTC)[reply]

Thanks, that new page is an excellent resource. One tiny request... if I relist an AFD, I still have to load the entire logpage to remove the afd, so it still takes several minutes.... lately almost 3 minutes??? I don't know why these particular pages take so long... but if you could provide an "(edit log)" link at the date headers (hypertext link with "&action=edit" suffixed) it would really help speed up the process of relisting afd's. Thanks, JERRY ^talk _contribs 23:44, 22 January 2008 (UTC)[reply]

I did that. I hope it helps with AfDs. Thank you for your work with that. Oleg Alexandrov (talk) 08:25, 23 January 2008 (UTC)[reply]

Thanks a lot. That's perfect. JERRY ^talk _contribs 12:33, 23 January 2008 (UTC)[reply]

Hello, please see this edit; apparently the bot is changing the name of Violetta Blue back to "[[Violet Blue (porn star)]]". This isn't correct; the porn star has changed her name to "Violetta Blue" because Violet Blue (author) brought a trademark infringement lawsuit against the porn star. "[[Violet Blue (porn star)]]" is a redirect page. Can you fix it so that it updates as Violetta Blue? Can you explain why this is happening? Thank you! Photouploaded (talk) 13:31, 11 December 2007 (UTC)[reply]

My guess is that since the biography project is so long, the bot was operating with information that was one or few days out of date. Let's see if it moves the article to the proper name next time it runs in a few days. Oleg Alexandrov (talk) 15:56, 11 December 2007 (UTC)[reply]

OK. Photouploaded (talk) 15:59, 11 December 2007 (UTC)[reply]

In [6] the bot added a category that was already present via the template. --NE2 20:30, 11 December 2007 (UTC)[reply]

Good observation. But the bot is not smart enough to detect when a template has the cat already, so it puts it in always. I'd think there is no harm to have the category show up twice (and most templates do not have the category I think). Oleg Alexandrov (talk) 03:31, 12 December 2007 (UTC)[reply]

Couldn't it just check for Template:cat importance or Template:cat class? --NE2 18:15, 12 December 2007 (UTC)[reply]

Right. I am reluctant however to make changes to that algorithm, that for two reasons:

• The way things are now there is no harm done adding the category second time (and there is no hard guarantee Template:cat importance will keep on having the category attached).
• I am trying to keep the code as free as possible of scenarios specific to individual language Wikipedias. The code (in one form or another) is already used at two other wikis, and may be used in more, and adding language-specific features without an overwhelming reason would make maintanance harder.

(Oh, and there could be a third reason, I'm just lazy :) Thanks for pointing out the issue though. Oleg Alexandrov (talk) 05:23, 13 December 2007 (UTC)[reply]

Thank you for your help. I'm new at editing Wikipedia articles, and you made my pictures look much nicer. I will change them to png, but the higher resolution will have to wait for now, I don't have an easy way of doing it without remaking the picture. I've actually made the changes to the Random Walks site as a school project for probability. I appreciate the help! —Preceding unsigned comment added by Whiteglitter79 (talk • contribs) 13:08, 13 December 2007 (UTC)[reply]

Let's see... User:Oleg Alexandrov, User:Jitse Niesen, User:Michael Hardy, User:Geometry guy, User:David Eppstein, User:Arthur Rubin, User:CBM, User:Fropuff, User:Charles Matthews, User:Paul August, User:Mikkalai. Soon to be User:Salix alba. Quite a hefty number if I do say so myself. --Cronholm¹⁴⁴ 17:25, 13 December 2007 (UTC)[reply]

Yep. There is also Lethe, R.Koot, and CSTAR (they are not that active recently). You'll make a fine admin too, in few months (you'd be ready even now I think, but given the politics of the place I;d say it is usually better to be around say for a year). Let's see how Salix alba's nomination will go. Oleg Alexandrov (talk) 04:29, 14 December 2007 (UTC)[reply]

Wow, Lethe, he's pretty much evaporated, I wish I could of interacted with him before that because he seems like a good guy. I think Salix's nom will go smoothly, voters seem to like candidates with similar profiles, smart, civil, and articulate. (David's and G-guy's RfAs comes to mind). As for me, I have pretty high standards for adminship, and I'd really like to do some more article work and admin related tasks before I even attempt at something like that, (and hopefully that work will concluded by the one year mark) but thanks for thinking I will do a good job :). Cheers —Cronholm¹⁴⁴ 15:02, 14 December 2007 (UTC)[reply]

I haven't quite evaporated. Yet.--CSTAR (talk) 06:09, 15 December 2007 (UTC)[reply]

Oops. :) Oleg Alexandrov (talk) 06:16, 15 December 2007 (UTC)[reply]

If you look at Wikipedia:Requests_for_adminship/Salix_alba it has already date stamped when it will end (the 20th), ideally that could be changed so the clock doesn't start until he writes in that he accepts the nom. and we can all start voting (unless we can start now?) Merkinsmum 14:11, 16 December 2007 (UTC)[reply]

Once Salix alba accepts the nomination, he will change to time to a week from that time (at least, that's what I told him on his talk page). Oleg Alexandrov (talk) 05:23, 17 December 2007 (UTC)[reply]

It should be a great nom. (He was top of my list of people to nominate.) Please don't vote until Salix alba accepts, and then, I agree, the date stamp needs updating. Geometry guy 20:15, 16 December 2007 (UTC)[reply]

(PS: actually, I stopped by for another reason, just to ask whether Michael missed the point of your comment at the round table, and if so, whether you think it was deliberate irony or accidental ;-)

He did miss the point, and I tend to think it was not irony. Oleg Alexandrov (talk) 05:23, 17 December 2007 (UTC)[reply]

Don't forget to change your bot now that the ArbCom elections are over. I protected the vote count page for a week so it won't keep logging useless edits. Grandmasterka 07:53, 17 December 2007 (UTC)[reply]

Thanks, I stopped the bot. Oleg Alexandrov (talk) 16:16, 17 December 2007 (UTC)[reply]

An editor has {{PROD}}ded the article about me for deletion. I think that this is improper, as it previously had a PROD removed, but I'd like a 2nd opinion. As I understand it, it's improper for me to contest a proposed deletion, per WP:COI. (The {{fact}} tags added are nonsense; they all come from the page used as source for that paragraph, but I'm sure it would be improper for me to remove the tags.) I'd like a propriety check. — Arthur Rubin | (talk) 09:35, 17 December 2007 (UTC)[reply]

I took care of the Prod tag. It is outside policy to prod articles that have survived Afd. Your article talk page has also attracted three sockpuppets. It appears you may have made some kind of wiki-enemy. Any thoughts on who and why? —Cronholm¹⁴⁴ 10:04, 17 December 2007 (UTC)[reply]

Well, there's the "41 vandal"; Abuse truth and Daniel Santos on "Satanic ritual abuse" and other articles; Levine2112 (unlikely) and allies (more likely) in Quackwatch.com; WAREL; the annivarsarists (adding anniversaries to future year articles); the century/millennials (who think that centuries and millennia begin with 00/000 rather than with 01/001); etc. It's hard to determine. It may have even been over a single reversion. — Arthur Rubin | (talk) 13:53, 17 December 2007 (UTC)[reply]

Thanks Cronholm. I do agree that the tag was improper, with the article having survived afd. Arthur, it appears to me that a band of sockpuppets want to give you some bad blood. :) I put it on my watchlist too, to see if the same trick is attempted in the future. Oleg Alexandrov (talk) 16:25, 17 December 2007 (UTC)[reply]

I don't know how mathbot works, but he deleted a link to Geometry and Topology even though it is in [[Category:Mathematics journals]] which is on List of mathematics categories. It might be a case sensitivity issue. There is a completely different article Geometry and topology. His own idea of his changes are at User:Mathbot/Changes_to_mathlists in the section December 17, 2007. JackSchmidt (talk) 20:20, 18 December 2007 (UTC)[reply]

P.S. These lists (including User:Mathbot/Changes_to_mathlists) are great. Thanks for this.

There are a few hyphenated articles that may have gotten lost too, but I've asked the hyphenator to check. In case this is manual, C P Ramanujam recently moved to C. P. Ramanujam via User:R.e.b. and User:Michael Hardy.

I assume it would be insanity to have restricted versions like Algebra or Representation theory? JackSchmidt (talk) 20:32, 18 December 2007 (UTC)[reply]

The bot will automatically reflect page moves. Let's see if it does that tomorrow.

Good point about Geometry and topology vs Geometry and Topology. A while ago I thought it would be good to not distinguish articles that differ only by case, I now changed my mind. Hopefully tomorrow after its run both articles will be listed by the bot.

Yesterday the bot had a bug due to a recent change I made, hopefully tomorrow things will come back to normal. Let me know if there is anything else which is odd. Thanks for your comments! Oleg Alexandrov (talk) 04:27, 19 December 2007 (UTC)[reply]

Everything looks good now (after the recent run), thanks!

I might have some questions for somebody this week about mediawiki guts. I basically want the recent page history of a few thousand articles on demand from javascript, but with the least amount of stress on the wiki foundation. Basically I want something like include/SpecialRecentchangeslinked.php = Special:Recentchangeslinked = Related Changes to spit out xml/json on query.php/api.php, but I don't think it currently can.

You (or your bot?) already have something functionally similar on tools.wikimedia.de, which is why I mention it to you. The question will likely be: which causes less wiki foundation stress, (1) hammering query.php with a bunch of (like 2000) page history requests, (2) hammering Special:Recentchangeslinked with a smaller number (more like 20) of target pages, or (3) submitting a patch to api.php/query.php that is more or less a copy paste inclusion of include/SpecialRecentchangeslinked.php and hassling someone to install it. I tend to think they are ordered from most stressful to least, but I wasn't sure about 2 vs. 3.

The question might be entirely different since I haven't read much about bots yet. My end goal is more of a human/client side tool, but it seems conceivable a little cgi or bot or something could do it. A simple bot that {{subst}}'d the 20 pages into 1 page every night would let Special:Recentchangeslinked do its work, for instance, but for all I know someone frowns on pages with 2000 links.

At any rate, thanks for the bot service (a scan of this talk page indicates it helps me more than I know). JackSchmidt (talk) 04:46, 19 December 2007 (UTC)[reply]

From your comments, I gather you know more than me about these things. :) If you want the recent history of a few thousand pages only occasionally, none of these approaches will cause trouble I think. I would go for the first, using query.php, I think that's cheaper than recent changes as the server gives you raw data rather than formatted html (if I understand things right). Just leave a second or two between subsequent requests, that will drown in the noise of the hundreds/thousands of queries per second. At a rate of one query per second, you'll get what you want in an hour or two, I think.

Pages with 2000 links are not a problem. More than 5-10 thousand simply won't work, the server request will time out before it displays it I think. Oleg Alexandrov (talk) 05:22, 19 December 2007 (UTC)[reply]

The Mathbot-generated edit counts that normally are on the RFA talk page seem to be missing. -- John Broughton (♫♫) 23:26, 18 December 2007 (UTC)[reply]

The bot was set up to not post the edit count unless the talk page of a given RFA is blank. I realized now that this condition is too restrictive. Now the bot will post the edit count unless the talk page already has the words "edit count" which would mean that somebody else did the job before the bot. Thanks for pointing this out! Oleg Alexandrov (talk) 03:56, 19 December 2007 (UTC)[reply]

I need your bot to organize Wikipedia:Version 1.0 Editorial Team/Alaska articles by quality statistics for Wikipedia:WikiProject Alaska. Any help in getting it to do that would be greatly appreciated! --Mr.crabby (Talk) 01:33, 19 December 2007 (UTC)[reply]

See if the instructions help. If you encounter any problems, please let me know. Oleg Alexandrov (talk) 03:49, 19 December 2007 (UTC)[reply]

Okay I got it working, thanks so much! --Mr.crabby (Talk) 22:23, 19 December 2007 (UTC)[reply]

Do you know if there is a way to generate graphics with textual commands, e.g. could I type \draw(0,0)--(1,1) or something like that and get what I want? or must a picture always be uploaded? Triathematician (talk) 01:23, 20 December 2007 (UTC)[reply]

My guess is that a picture must be always uploaded. LaTeX supports to some extent what you say, but I think the subset of LaTeX used on Wikipedia does not have that. Oleg Alexandrov (talk) 04:29, 20 December 2007 (UTC)[reply]

Oleg, wheel-warring is strongly discouraged, particularly when an admin deletes something on BLP grounds. I would appreciate it if you would re-delete that edit summary (which was gratuitous and provocative, and accompanied by posting the same link twice within one comment), then discuss the issue with me by e-mail. SlimVirgin ^{(talk)(contribs)} 11:33, 20 December 2007 (UTC)[reply]

Oh, wheel warring is surely a bad idea, and only generates acrimony. Wheel warring is defined when administrators repeatedly undo each other's actions, which, if I may note, did not happen.

You deleted a version from a talk page history without explanation, which is an improper use of admin tools. I restored it, since policy discourages such actions except only when absolutely necessary.

And you keep on referring to the BLP grounds, even though the discussion at WT:BLP was unanimous in that BLP does not support your actions at that article.

So let me ask you again what I asked you several times, please explain, in the face of everybody else thinking to the contrary, what is wrong with the link at talk:Carl Hewitt. Thank you. Oleg Alexandrov (talk) 15:32, 20 December 2007 (UTC)[reply]

Howdy I un-reverted an anonymous edit to Orthogonal group that you reverted. The anon editor had actually caught a mistake in the article, but their edit was not optimal so I think your revert was sensible. I am leaving a note in case you have a contradictory source that indicates both orthogonal groups of dimension 4 over the field with two elements are exceptions (I have a source or two that says no, but the situation is complicated, and I would not be surprised to find a contradictory source). If you were just slowing the flood of number tweaking vandalism that this looked like, then feel free to ignore the rest.

I made a slightly better attempt at correcting the error (a few of us are preparing to rewrite the finite geometry sections next year, but right now so much is missing it is hard to be strictly correct). I gave a citation for the true statement, and tried to indicate in the article what the true statement was, but lots of other true and nearly true statements sound similar, so I wanted to mention it in case you had checked elsewhere and found a contradictory source. I mean, what does "reflection" even mean in characteristic two? Fix a hyperplane and send an orthogonal vector u to −u ... except that is called the identity matrix in characteristic two. Another wikipedia article defines reflection to be any element of order two, but then all orthogonal groups are generated by their elements of order two, and so there is no exception. I took the definition that is basically in Groves (citation on the article), that is, a reflection in characteristic two is a symplectic transvection that happens to be orthogonal (since orthogonal is contained in symplectic in characteristic two). JackSchmidt (talk) 21:28, 23 December 2007 (UTC)[reply]

Sorry, at first sight I thought that the comment about "witt-index" was uhm, a naive attempt at wit. Feel free to change the article in any way you feel correct of course. Thanks for catching that! Oleg Alexandrov (talk) 22:23, 23 December 2007 (UTC)[reply]

Hi, this GA falls under Wikipedia:WikiProject Textile Arts but your bot neglected to add it to our project list. And after I added it manually the bot removed it. It's a small WikiProject and we need the few feathers we have in our caps. Please advise? Thanks, Durova^Charge! 12:07, 28 December 2007 (UTC)[reply]

From what I see at Talk:History of silk, Category:GA-Class Textile Arts articles is a redlink, so the bot can't see it. Try to see if following the instructions on how to set up the bot and run it right away would help. If not, please let me know. Cheers, Oleg Alexandrov (talk) 03:45, 29 December 2007 (UTC)[reply]

Ah, I think I get it. Okay, I've adapted the text from our FA category and started the GA category. Thank you very much for the tips. Happy New Year, Durova^Charge! 04:09, 29 December 2007 (UTC)[reply]

The Original Barnstar I figured I'd give you this for creating the very useful WP 1.0 bot since I used it a lot of the time in regards to the projects I'm part of. Well done and keep up the great work! -Animagentile (talk) 05:32, 31 December 2007 (UTC)[reply]

Thanks. :) Oleg Alexandrov (talk) 15:44, 31 December 2007 (UTC)[reply]

I saw in my watchlist that Oleg said "thanks", and wondered "Hmm, Oleg is usually the one to help everyone and get thanked, I wonder what this is about". Well deserved, no doubt - though I'm personally not sure what WP 1.0 bot does exactly :-). -- Meni Rosenfeld (talk) 16:47, 31 December 2007 (UTC)[reply]

This reminds me of when I said "thank you" to one of my high school teachers after she said "thank you" to me. She reminded me that what I should have said was "you are welcome". Since the barnstar is a form of thanks, I think the proper response would have been to say "you are welcome". JRSpriggs (talk) 03:39, 1 January 2008 (UTC)[reply]

OK, your're all welcome! :) Happy New Year! Oleg Alexandrov (talk) 04:57, 1 January 2008 (UTC)[reply]

Long time, no see! You're probably wondering who I am, right? When I first joined Wikipedia in mid-2006 (I believe it was May or June), you were the first person to cordially welcome me, I think. It was such a long time ago I forgot! I saw your name, and it looked very familiar, so I just wanted to drop by, say "Hi", and thank you for doing that kind gesture! — Cuyler91093 - Contributions 05:47, 5 January 2008 (UTC)[reply]

Heh, you're welcome. :) Oleg Alexandrov (talk) 06:59, 5 January 2008 (UTC)[reply]

"I.Q. (film)" appears in the List of mathamatics articles. —Preceding unsigned comment added by Dr. Api (talk • contribs) 22:31, 7 January 2008 (UTC)[reply]

I would guess that this is because the article is in Category:Mathematical films. Probably, it should not be, but I have not seen the film so I am not certain. JRSpriggs (talk) 04:04, 8 January 2008 (UTC)[reply]

Yes, that's due to the categorization. I will now remove the math films category as this film is only periferially mathematical. I will also remove the article from the list of mathematics articles. Thanks for pointing this out. Oleg Alexandrov (talk) 06:53, 8 January 2008 (UTC)[reply]

Hi, your bot hasn't updated Wikipedia:Version 1.0 Editorial Team/Cornwall-related articles by quality since the 2nd January - is there a problem with it? Thanks DuncanHill (talk) 01:41, 8 January 2008 (UTC)[reply]

The bot appears to be running. Perhaps it just had a bigger backlog and could not run fast enough. If it does not get to it in a couple more days, then please let me know. Thanks. Oleg Alexandrov (talk) 06:57, 8 January 2008 (UTC)[reply]

Ah, I actually found what was going on. Three instances of the bot were running, and stepping on each other's feet rather than helping each other. I now killed two of them, and the third one will do the job. Thanks for noticing this, as some projects did not update even since December 30. Oleg Alexandrov (talk) 07:03, 8 January 2008 (UTC)[reply]

Thanks for taking care of this. Too bad the bot has to start all over again with the projects starting with A. --Descendall (talk) 07:38, 8 January 2008 (UTC)[reply]

Cool, thanks for your help, best wishes, DuncanHill (talk) 11:12, 8 January 2008 (UTC)[reply]

It still hasn't been updated - but some others have been updated twice since your message on the 8th. Any ideas what's happening? DuncanHill (talk) 03:24, 12 January 2008 (UTC)[reply]

Your bot has updated Wikipedia:Version 1.0 Editorial Team/Alternative music articles by quality twice in the past 16 hours, but it hasn't updated Wikipedia:Version 1.0 Editorial Team/Peru articles by quality log‎ once in the past 10 days. Do you have any idea when the bot will get around to updating WikiProject:Peru? Or will it just never get there because it goes back to the beginning of the list at the end of the day? --Descendall (talk) 05:33, 9 January 2008 (UTC)[reply]

The bot had some glitches, but is now running and will do the job, hopefully in a day will get to Peru. I'll make sure the bot runs more seldom from now on, covering all the projects roughly the same amount. Oleg Alexandrov (talk) 05:57, 9 January 2008 (UTC)[reply]

Re-reading what I wrote, I realize that you could interperate it to be kind of snotty. That wasn't my intent. It's hard to tell someone's tone of voice over the internet. Anyway, thanks a lot. I don't know what we'd do without that bot. --Descendall (talk) 08:10, 9 January 2008 (UTC)[reply]

No, it did not even come to my mind to not interpret it that way. (Plain text is notoriously hard to interpret, I've learned for a while that when reading it is usually best to stick to the facts presented than try to think of the intent of the person who wrote. So, if you want to piss me off you'll have to be explicitly hostile, e.g., "your idiot bot is Peru-hater", or something. :) Oleg Alexandrov (talk) 16:15, 9 January 2008 (UTC)[reply]

I fixed one more glitch. The bot will get to Peru in less than half a day, I promise. :) Oleg Alexandrov (talk) 17:11, 9 January 2008 (UTC)[reply]

But it still has not reached Yorkshire where the last auto BOT run was 2nd January! Keith D (talk) 12:10, 11 January 2008 (UTC)[reply]

The bot is running. I hope it will get to Yorkshire in a few days. The recent glitches are something rare, before then the bot ran without any problems for at least a couple of months. Let's hope that with the change in running times I made recently it will start running glitch-free again (you can run the bot by hand if you want, in the meantime). Oleg Alexandrov (talk) 16:11, 11 January 2008 (UTC)[reply]

"Your idiot bot is Peru-hater." Just kidding. But it did skip the Peru articles when it did the Ps. The Peru Project is just barely active, so it's not that huge of a deal, but I thought you might want to know that the thing is apparantly skipping some projects. --Descendall (talk) 20:10, 17 January 2008 (UTC)[reply]

The bot is still working on some projects it did not get to since last year. It will get to Peru towards the end of the run (that is to say, the bot does not go chronologically, but rather in the order of first doing the projects which were not done for a while). If you want, you can run the bot by hand as specified at the instructions (assuming that the Peru project is not too big). Oleg Alexandrov (talk) 04:44, 18 January 2008 (UTC)[reply]

I concur with the U.S. roads projects - most (spread through the alphabet) have not been run since January 3. --Rschen7754 (T C) 04:46, 12 January 2008 (UTC)[reply]

The bot will start a new run in a couple of hours. From then, all projects will be reached in around four days. Sorry I can't be more helpful, we're still reaping the results of the bot glitches around the new year, which I hope won't happen again soon. Oleg Alexandrov (talk) 21:41, 12 January 2008 (UTC)[reply]

Oleg, I propose to update the Solid angle page with the section displayed in my Sandbox - user:Fjackson/Sandbox. However there is a conflict with vector notation. A previous user has used an "arrow above" the letter rather than an "underline below" the letter to indicate a vector. What is best way forward? Please fell free to amend my proposed update. Frank M Jackson (talk) —Preceding comment was added at 19:22, 9 January 2008 (UTC)[reply]

I sort of think that think an arrow on top is a more standard notation for a vector. But what matters most is that the article notation be consistent, so if you want to expand that article and use the underline below notation, I think that is OK, as long as the new notation is used everywhere in the article. 16:09, 11 January 2008 (UTC)

I'm just curious - what's the main source of delay in the bot, that makes it run so slowly? Is it the editing rate, or the download rate, or something else? It seems remarkable to me that it requires three days to complete - how many edits per run does it make? — Carl (CBM · talk) 03:26, 14 January 2008 (UTC)[reply]

The bot does one query every 1 second and an edit every 5 seconds. I'll be able to answer the question about how many edits it makes after I get a complete untampered log for a recent time, which would be in five days maybe. Oleg Alexandrov (talk) 16:31, 14 January 2008 (UTC)[reply]

I'm not sure if you know about the "maxlag" method of rate limiting.[7] The idea is that you send a parameter with each edit, and if the server lag is too high the server tells you to wait and try again. But if the lag is not too high, there is no need to add an extra delay after each write anymore. I have written some perl code to do this (I needed extra functionality compared to perlwikipedia); you might want to look at it.[8]. The watchlist part seems to be buggy, and the output is ugly, but the editing and maxlag parts work. It's based on some code from my API library.[9] — Carl (CBM · talk) 03:04, 15 January 2008 (UTC)[reply]

Continued at WT:1.0/I#Why is the bot slow?. Oleg Alexandrov (talk) 06:11, 15 January 2008 (UTC)[reply]

14-Jan-2008: Oleg, can you make the WP Bot not wikilink the dates/years of the "Date" column in all those status files? From each status WPBiography file, listing 400 articles, I have been unlinking the remaining 75% of dates/years (about 600 wikilinks unlinked per each status file).

The format change would be: replace "[[May 5]], [[2007]]" as "May 5, 2007" for all 400 dates per file. I have been editing each file by hand, to also fix some "Talk:.../Comments" pages, but I suspect the 600 unlinked dates/years in each file will be relinked once WP Bot processes those same article names. One of the edited files has been (version 1-Jan-2008):

• Wikipedia:Version 1.0 Editorial Team/Biography articles by quality/34

That Biography quality group "34" had nearly 600 overlinked dates/years, which is about the average, for a total effect: 600 links * 110 status files = 66,000 dates/years overlinked.

There's no hurry on this, because many other WikiProjects are adding millions of overlinks in their propagated navboxes: well over 50,000 different articles are linking far many more words/dates than needed. The date/year linking by WP Bot is just a tiny portion of the total impact. Thanks. (I'll check back here for a response in a few days). -Wikid77 (talk) 04:41, 14 January 2008 (UTC)[reply]

Well, fixing the dates by hand knowing that the bot will overwrite them was probably not a good idea. :) I will get to this sometime this week. Good idea, I don't like overlinked dates either. Oleg Alexandrov (talk) 16:30, 14 January 2008 (UTC)[reply]

Done. I actually did this fix a long time ago, what I did now was remove the links from the dates which were linked to before that fix. At the next run the bot will remove all links from dates. Oleg Alexandrov (talk) 04:51, 22 January 2008 (UTC)[reply]

This bot hasn't updated the information on the page for over a week. We assessed all of the articles in the project. What could be the problem? miranda 18:11, 14 January 2008 (UTC)[reply]

I ran the bot by hand using the instructions for WP:FRAT (which can be done at any time in the future you need a run right away). About the more general issue of the bot lag, hopefully now this is fixed and the bot is doing its job. Oleg Alexandrov (talk) 06:01, 15 January 2008 (UTC)[reply]

Thanks for your help! miranda 01:57, 16 January 2008 (UTC)[reply]

Would you be interested in providing a clone of WP1.0 bot to oversee the Australia WikiProject. A discussion is currently taking place at Wikipedia:Australian Wikipedians' notice board#Wikipedia:Version 1.0 Editorial Team/Australia articles by quality log with concerns that although the bot is performing some outstanding work, it's taking a long time to get back around to us. Whether we settle for the bot work to continue as is, a clone of the current bot, or a new bot capable of supporting our project needs as a whole is still under discussion. -- Longhair\^talk 06:16, 15 January 2008 (UTC)[reply]

These days the bot is kind of slow, but this is an exception (and this is being addressed). Usually it's been (and will run) once every four days. Is that still too slow overall? Oleg Alexandrov (talk) 06:17, 15 January 2008 (UTC)[reply]

I remember the days of 24 hourly updates :) I don't mind the delay, one can always have the bot run manually if there's a pressing need for a new log, however some editors are finding other uses for the log and would like to see it updated more often than twice weekly. I understand our project size causes a considerable delay to the bot in getting to other projects, which has brought us to thinking a dedicated bot may be the way ahead? -- Longhair\^talk 06:24, 15 January 2008 (UTC)[reply]

The recent delays are partly because I have too little time to supervise the bot, so unfortunately I don't have the time to supervise another instance of it (sincerely sorry). If however, somebody wants to get a copy of the code (available from the bot page), and host an instance of it on some server, that will be perfectly all right with me (I would be very grateful too). Let me know if any of the Australia wikiproject members could do that. Oleg Alexandrov (talk) 06:37, 15 January 2008 (UTC)[reply]

How much work is there in supervising the bot? I'd be willing to host the bot given that it's not an around the clock task and something that merely requires a restart now and again? -- Longhair\^talk 06:56, 15 January 2008 (UTC)[reply]

That's right, doing restarts and listening to complains when the bot lags is all there is to it. :) Detailed installation instructions are available from the bot page. I can help if you encounter any issues. Cheers, Oleg Alexandrov (talk) 06:58, 15 January 2008 (UTC)[reply]

I've worked with perl in the past and can follow the instructions ok for now. I'll setup a clone of the bot and take things from there. No doubt you'll be soon hearing from myself again ;) Thanks for your quick and helpful replies. -- Longhair\^talk 07:01, 15 January 2008 (UTC)[reply]

That's fine by me. My memory is that the deletion was to do with the attention he was getting at that time. SlimVirgin ^{(talk)(contribs)} 16:00, 16 January 2008 (UTC)[reply]

There might be a slight error here in article Pinsky phenomenon Instead of < there should be <= . See Pinsky's book, around p. 142+ about this. Maybe you can correct. Wikipedians editors do not like me and call me names and post stuff on my talk page. They frighten me and make me sad. Maybe you can help instead of me. AmeliaElizabeth (talk) 12:04, 17 January 2008 (UTC) AmeliaElizabeth (talk) 12:08, 17 January 2008 (UTC)[reply]

I wrote this to an editor who keeps harassing me. Can you help me?

why do you not like me? why are you so mean? why do you make me sad?

I mean no harm. I love numbers. Gauss proved all numbers interesting. Why are you mean to me? Why you make me sad? I am new here. I thought people are supposed to be nice. why not nice? please explain. AmeliaElizabeth (talk) 12:16, 17 January 2008 (UTC) AmeliaElizabeth (talk) 12:19, 17 January 2008 (UTC) AmeliaElizabeth (talk) 12:24, 17 January 2008 (UTC) AmeliaElizabeth (talk) 12:24, 17 January 2008 (UTC)[reply]

Mmm, how can I help you? I took a look at your talk page and some of your edits. I happen to agree with the folks who reverted this edit you made. This edit was not so helpful either. If somebody's bugging you too insistently, please let me know. I'd suggest though that you follow the advice you are given on your talk page about not adding questionable material, it probably does not improve the articles much.

About Pinsky phenomenon, I guess you can do the fix yourself. You are the original article creator, and you should know best what it is about. Cheers, Oleg Alexandrov (talk) 15:48, 17 January 2008 (UTC)[reply]

you may be interested in Wikipedia:Requests for checkuser/Case/Alfred Legrand. CM (talk) 18:50, 18 January 2008 (UTC)[reply]

Oh. :) Oleg Alexandrov (talk) 04:16, 19 January 2008 (UTC)[reply]

I was wondering why the bot does not recognize and categorize FL-Class in its assessments? Is there anyway that it could be written into the bots code to have it categorize FL-Class and "FL-Class Foo articles" and place those in the statistics. The thing is that we now have around 600 Featured lists and seem to be promoting around 35 a month, so in a little bit it will rival FA-Class article in size. I was just wondering and thanks in advance.
Gonzo fan2007 ^{talk ♦ contribs} 21:32, 18 January 2008 (UTC)[reply]

Try raising this at WT:1.0/I. The primary problem is not technical, I can easily add the new class. It is just there seems to be no agreement to add it. Maybe it is time to restart the discussion. Oleg Alexandrov (talk) 04:17, 19 January 2008 (UTC)[reply]

Your input on the ideas presented at Wikipedia talk:WikiProject reform#Single Banner? would be very much appreciated. Thank you. John Carter (talk) 18:09, 19 January 2008 (UTC)[reply]

Howdy. I added a Category:Topical indexes cat-tag to the 27 "List of mathematics articles (x)" subpages (eg), but mathbot undid that when it updated. I was wondering if you could add that category in to mathbots default output. Thanks :) -- Quiddity (talk) 21:08, 20 January 2008 (UTC)[reply]

I added the category to "List of mathematics articles (x)". Cheers, Oleg Alexandrov (talk) 05:25, 21 January 2008 (UTC)[reply]

I spent a little while tonight working on the WP 1.0 bot. It wasn't hard to implement what you suggested to fetch revision ids in larger batches, but I don't think that will help a lot, since they only seem to be fetched then the quality of an article changes.

What does look like it might help is to fetch the data stored in the server assessment pages more efficiently. It's possible to fetch the content of multiple pages at a time, rather than one at a time. For projects like Australia that have over 100 assessment subpages, this would significantly decrease the number of page requests.

I have the code to do the downloading. The issue I have is the "breakpoints" that you parse and use. What is the purpose of those? Won't the code automatically split the data into subpages? — Carl (CBM · talk) 03:28, 25 January 2008 (UTC)[reply]

The purpose of breakpoints is, to an extent possible, not split the data into subpages anew each time. For example, if say there was a list of 1000 articles in 5 pages of 200 each, and the first article got removed, I'd rather have 199 articles in the first page and not modify at all the other ones.

Breakpoints are used indeed only to parse existing pages. My own guess is that the most bot effort is spent getting new data, that is, reading the articles from the categories and the history ids of pages whose assessment changed.

That is to say, the bot each time reads anew the stub class biography category, all 300,000 of them, in batches of 500.

And there can be a lot of articles whose quality changes too.

Anyway, if you want, I can give you write access to the SVN repository if you are ready to merge in your changes (you'd need to tell me a gmail user account). Thanks a lot for your help! Oleg Alexandrov (talk) 04:17, 25 January 2008 (UTC)[reply]

Thanks for the offer. I spent a little while thinking about it this morning.It seems to me that, since the current code is stable and works, there is no real emergency rush to improve it. We should consider the "right way" for it to run, not just the way that the current code has evolved.

• The current code works. Its main problem is the large number of requests that must be made. The cause for all these requests is the large volume of data that is used.
• Incremental improvements, such as grouping together oldid requests, will speed up the bot some, but as the amount of data increases, the bot will once again take too long to run.

It seems to me that there are at least two options that will address the root cause of the slowness, which is the enormous volume of data. One is to make the bot multithreaded so that it can update several projects in parallel. The other option is to change the way the bot gathers data, so that it doesn't use the API. It would be possible to gather the data you are looking for with a much smaller number of SQL queries. For example, all the rating information for an entire project could be fetched in one query from the toolserver's database replicas, rather than dozens of queries to api.php.

What are your thoughts on the matter? — Carl (CBM · talk) 16:32, 25 January 2008 (UTC)[reply]

Well, making the bot multi-threaded will just add more strain on the Wikipedia server I think (I could as well run several instances of the current bot in parallel, spaced by one day). Using database replicas could be problematic I think, since sometimes those tend to be days (if not months) behind, as far as I remember (if we two are talking about the same thing).

I think any incremental improvements would help. More than 1.3 million of the current 2.2 million Wikipedia article are already reached in some way by the bot. The rate of growth won't be that large in the future I would think.

So, if you can make some improvements which don't require much work but can make the bot much faster that would be awesome.

If, on the other hand, you'd like to write a complete replacement for the bot, me, and everybody else, would be awesomely grateful. Oleg Alexandrov (talk) 16:46, 25 January 2008 (UTC)[reply]

I'll start by fixing the oldid download problem, so that they are downloaded in larger chunks.

I wasn't thinking of the database dumps, which do get outdated. The toolserver cluster has a SQL server that replicates large parts of the enwiki database in real time. There is some replication lag, but usually no more than a few minutes, which would be fine for our purposes.

My impression from various places is that server load is not as much of an issue as it was once made out to to be. Running four instances of the bot, spaced out by one day each, is actually a very nice idea. The load the bot creates is really not very high, so I can't see why four instances would cause any problems. With the web form, you already might have four instances running at the same time. — Carl (CBM · talk) 16:55, 25 January 2008 (UTC)[reply]

I remember times when the replication lag was weeks. :)

I can't run that many instances of the bot since each of them can use as much as 1GB of data, and the memory of the server is only 3GB (shared among other users too). I can run a bot every other day, but I wonder if there is really a need for it, I would think that the current four-day run would suffice (or wouldn't it?).

Looking forward to getting your fixes.

And if you want to create a code which would do things by quering the database directly, that would be great (my own code can be kept as backup for cases when the database lag is huge). We could also have it host its pages on an outside server, bypassing the need for editing wiki pages. That way one could do a complete run in a matter of hours I think.

Cheers, Oleg Alexandrov (talk) 04:11, 26 January 2008 (UTC)[reply]

I have made quite a bit of progress on the script, to the point where I was able to make a test run yesterday with the editing disabled. Now I need to review the typescript to look for any errors. This post ended up being somewhat long, sorry about that.

All the projects except biography ran successfully in about 7.5 hours (with no editing). Here are the stats from the run (including reading the biography project data but not trying to upload it):

• 19652 HTML requests for category contents, page content, and oldids
• 1413 more HTML requests to get the correct name of an article
• 8318 page writes (attempted but disabled)

At 5 seconds per edit, those edits would be about 11 hours. So I conservatively estimate I can get the running time under 36 hours.

The bot was able to run quickly because server lag was low. The way it is set up, instead of being throttled by sleeping between requests, the script is throttled by the server lag (the maxlag method I mentioned).

I didn't try to disable the biography project, but I learned by experience that the machine I'm using for development doesn't have enough memory to run it. I left it run overnight, but it was just thrashing and didn't complete the project. The memory use was right around 400mb until the biography project, when it shot up to about 1.7gb (on a 64 bit processor). The increase happened during the part where the script is loading the old data from the server into memory. I don't think optimizing this is a high priority, though.

I registered the gmail account cbeckhorn at gmail.com to use with Google code. — Carl (CBM · talk) 16:39, 27 January 2008 (UTC)[reply]

Wow! I'll reply further by email. Oleg Alexandrov (talk) 17:03, 27 January 2008 (UTC)[reply]

Hi. Mathbot is currently not listing the open discussions of January 19th correctly (lists is at 0 open and 0 closed!). This may be due to some tag or template on the AfD day page of course, but I haven't been able to find it so far. Could you have a look? Fram (talk) 10:54, 25 January 2008 (UTC)[reply]

The html code generated by the server changed, and that confused the bot. I fixed it now. Thanks for pointing it out! Oleg Alexandrov (talk) 16:39, 25 January 2008 (UTC)[reply]

Hi Rabbit. Just a note. My view is that the picture at Differential equation looked better the way it was before you moved it. It was larger (so easier to see), and fit a hole in the TOC to the right. Can you explain why you moved it? Thanks. Oleg Alexandrov (talk) 07:14, 27 January 2008 (UTC)[reply]

Hi Oleg, I removed the explicit "size" parameter from the picture as that seems to be the Wikipedia official recommendation for most images. I moved it up to the top since it was making the table of contents wrap, and it made article navigation more difficult (for me at least). Feel free to move it back, but I would still like to keep the explicit image size out. The picture is not essential to understanding the article, just eye-candy basically. Silly rabbit (talk) 14:26, 27 January 2008 (UTC)[reply]

You are right about the official recommendation. But if an image can't be seen at the official recommendation, then I guess one may try to make it larger. :) I made the image 250px instead of the original 360px. Does it still wrap the TOC? Oleg Alexandrov (talk) 17:06, 27 January 2008 (UTC)[reply]

No, it's fine now. Problem solved! Silly rabbit (talk) 17:08, 27 January 2008 (UTC)[reply]

Hello Oleg,

Sorry for being kind of dumb, but I misspelled the name for an image I uploaded:

File:ResidueTheoremIllustratioin.png

could you please fix it? Also delete the older upload if that's possible, it's rushed and completely useless. —Preceding unsigned comment added by Ben pcc (talk • contribs) 17:23, 29 January 2008 (UTC)[reply]

Done. I moved it to commons as commons:Image:Residue theorem illustration.png, and made residue theorem point to the new image. Normally I'd copy the file history too, but the tool for exporting image history from Wikipedia to commons is broken today. Still, you figure as the image creator on the commons page. Cheers, Oleg Alexandrov (talk) 03:53, 30 January 2008 (UTC)[reply]

Thank you! -Ben pcc (talk) 05:33, 30 January 2008 (UTC)[reply]

I commented on Talk:Gauss-Newton algorithm that I thought the move of the sandbox to a new page was a good idea, but then I noticed that one of the good titles was already taken. Is this something I can fix without admin tools?

BTW I wanted to make sure my comments were not offensive. I wanted to get across the idea of a bold, expert written article versus a humble, clearly written article (hopefully useful for our new amazingly expert editor), and compared hypothetical editors as bold or timid. Hopefully it was clear I am not trying to comment on you as an editor (I explicitly named such a timid editor, me; I'm not about to spend hours making sure I *really* understood the statistical foundation of NLA), but since I wasn't merely voicing support for your opinions I wanted to make it clear.

Also, if you care, I definitely think least squares needs work. I just think that prior to the new expert's edits, that work also included more statistics :) The whole thing *still* needs to be written for the layman. JackSchmidt (talk) 19:05, 31 January 2008 (UTC)[reply]

I got your intention allright, don't worry. :)

Petergans prefers Non-linear least squares from what I see in the text, so I moved the article there and redirected Nonlinear least squares to it. We can discuss further at Talk:Non-linear least squares whether there should be a dash or not. Oleg Alexandrov (talk) 04:12, 1 February 2008 (UTC)[reply]

Hi Oleg.

I am not happy about your immediate reverting of my work. I did not use self-invented notation. The multiplication sign, · , was not invented by me, though I do not know who invented it. All programming languages need a multiplication sign. The differential sign, d , was not invented by me either, but by Leibniz. Einstein used ds² for the square of the metric. Not only did he use the differential, but he squared it! The procedure for formal differentiation, that I described, was used by Leibniz and Bernoulli before Weierstrass and the other limit guys.

Variable names instead of function names are very much used in physics. Using V for volume, P for pressure, T for temperature, S for entropy, and U for energy, the law of thermodynamics is dU = TdS−PdV, no matter which of the variables are considered to be the independent ones.

WP needs an elementary explanation on the formal use of differentials, which is inherent in the Leibniz notation. The fact that the differential dx has no interpretation as a limit does not mean that the differential must be banned. It is used formally in differential algebra. When you call the derivative a = dy/dx, and the integral y = ∫a dx, then you must explain to the readers why you do not dare to write dy = a dx.

Please remember that you are supposed to improve rather than to revert.

Bo Jacoby (talk) 19:03, 1 February 2008 (UTC)[reply]

The notation ${\displaystyle df}$ is used for differential forms. The notation ${\displaystyle f\cdot dx}$ is used for line integrals of vector fields. However, I have never seen the derivative being calculated as

${\displaystyle d(x^{3})=3x^{2}\cdot dx.}$

People use

${\displaystyle {\frac {d(x^{3})}{dx}}=3x^{2}.}$

If you disagree, please provide references to textbooks using your notation. Wikipedia is not the place for original notation. Per WP:NOR we must follow existing conventions.

And about me reverting your work, there was plenty of talk at Talk:Integral#Explicite multiplication signs to improve legibility for new readers which you failed to heed. So once again, please do not invent new notation. Oleg Alexandrov (talk) 03:42, 2 February 2008 (UTC)[reply]

I have frequently used expressions such as ${\displaystyle d(x^{3})=3x^{2}dx}$ as intermediate steps in calculations. I am sure that I was imitating others rather than inventing it, although (many decades later) I do not remember from which source I learned to do it that way. JRSpriggs (talk) 07:15, 3 February 2008 (UTC)[reply]

Thanks for answering!

Leland McInnes, commenting my suggestion at Talk:Integral#Explicite multiplication signs to improve legibility for new readers, confirms that "the sort of manipulations you are using were considered quite normal". So it is not original research on my part. The differential being calculated as "${\displaystyle d(x^{3})=3x^{2}dx}$" was once considered quite normal, even if you have never seen it before. The reason why it is no longer considered normal is that it has no interpretation as a limit, as I explained above. But still it is an elementary fact of differential algebra.

The expression "adx" is a special case of a differential form in one dimension, and the expression "∫adx" is a special case of a line integral in a one dimensional vector space. Notations applicable for general cases, such as "dy" and "∫a·dx" , also apply for special cases.

The multiplication sign is in many cases omitted, but sometimes it improves clarity. In the expression "5·5" it is indispensable, and it is not new notation. In a non-commutative ring, where "xy" is not equal to "yx", you cannot write "d(xy)=xdy+ydx", as you usually do, but you must conserve the sequence of the factors: "d(xy)=xdy+dxy". As the expression "dxy" can be confused with "d(xy)" the explicite multiplication signs in "d(x·y)=x·dy+dx·y" are indeed convenient.

The opposition at Talk:Integral#Explicite multiplication signs to improve legibility for new readers was not against explicite multiplication signs in general, but against using explicite multiplication signs in Riemann integrals, and I did heed that opposition, in that I did not introduce explicite multiplication signs in general in that article.

Please reconsider the legitimacy of your reverting my contribution. Bo Jacoby (talk) 16:32, 3 February 2008 (UTC).[reply]

The multiplication sign is not used except for things like ${\displaystyle 1\cdot 2}$ (and in the first grades in school). I have taught mathematics at good U.S. universities for eight years and I have never seen notation like ${\displaystyle 3x^{2}\cdot dx.}$ Please provide references to established calculus textbooks using that notation. Otherwise, we can't use that notation on Wikipedia. Oleg Alexandrov (talk) 16:39, 3 February 2008 (UTC)[reply]

Hi Oleg. OK, I accept the implicite multiplication. Can you, following User:JRSpriggs, accept the differential "${\displaystyle d(x^{3})=3x^{2}dx}$" ? If yes, then please undo your revertions and 'improve' my contribution by deleting multiplication signs. Bo Jacoby (talk) 20:51, 3 February 2008 (UTC).[reply]

Please raise the issue at Talk:Derivative if you wish. I am myself firmly against notation not supported by references. dx is not an infinitesimal, and should not be misused. It has its use in differential forms. And when doing separation of variables it is convenient to think of it as a variable. When differentiatiting functions however, the only estiablished dx notation is Leibniz notation, ${\displaystyle df/dx,}$ so I'd stick to that. Oleg Alexandrov (talk) 21:45, 3 February 2008 (UTC)[reply]

I'm not sure whether it's more convenient for you to communicate by email or on the wiki. I stopped my tests after about 24 hours. The editing took much longer than I expected; the bot seems to average under 3 edits per minute, which is far too slow to finish in under two days. I am running the regular assessment script on kiwix now, so that it can complete the run that I halted to do my tests.

I have a couple things to do now on my own machine before the next test:

• There was an official announcement on the mailing list that query.php is deprecated and may be removed soon. To that end, I need to switch the script to using api.php. They have ported the 'txt' output format from query.php to api.php, so this is not as difficult as it sounds.
• I want to switch to a new editing library that uses maxlag instead of a fixed delay. I have been using this library in some of my scripts for a couple weeks already. I hope it will speed up the editing process.

I'll let you know when I make progress in these areas. — Carl (CBM · talk) 14:41, 3 February 2008 (UTC)[reply]

Hi Carl. I forgot to check my old school email yesterday but was on Wikipedia frequently. So sorry, I think nowadays I can be reached faster on this talk page.

About installing the new XML library, I use the directory public_html/wp/modules for libraries. I usually use CPAN to install new libraries, but see what works for you.

I know query.php is deprecated, I did not know they planned to remove it outright. If you have the time to do the migration that would be great (I hope it won't be more than replacing "query" with "api" in the server request and some other minor syntax changes).

You're more than welcome to use the new editing library based on maxlag.

Thanks a lot! Oleg Alexandrov (talk) 16:45, 3 February 2008 (UTC)[reply]

Using this page is fine with me. I think the query.php replacement will be that simple, yes.

I have the unchanged script running right now, and it was on attempt 640 to upload Wikipedia:Version 1.0 Editorial Team/College football articles by quality/3. I started it again with a reduced limit on the number of attempts. I hope whatever error this is will be resolved by the switch to the new editing library; it doesn't seem like it's worth debugging unless you already know the cause. — Carl (CBM · talk) 03:52, 4 February 2008 (UTC)[reply]

I have no idea what the issue could be. Thanks for reducing the number of attempts, 1000 was too much anyway. I killed the bot again, and restarted it after uncommenting the line "&decide_order_of_running_projects...". That routine makes the project run not in alphabetical order, but in the order of first running the projects which have not been done for a while.

After restarting the bot I commented that line again as it causes problems when several instances of the bot (started several days apart) focus on the same project at the same time (that's what happened around the New Year).

It is good though you are aware of that subroutine, it is helpful when the bot gets restarted many times and you'd rather want it to continue where it left off than start at "A" again.

Thanks for looking at the bot! Oleg Alexandrov (talk) 04:14, 4 February 2008 (UTC)[reply]

I was aware of that subroutine, and noticed it was commented out in the production version, but didn't know why. It has a side effect that is very useful in debugging - if the bot has an error and I kill it to edit the code, it will return to the same point very quickly when I run it again. — Carl (CBM · talk) 04:18, 4 February 2008 (UTC)[reply]

Cool. :) II increased the number of attempts from 25 to 200 though. That is good when the database is down for a few minutes. I have no idea why 640 attempts were not enough, I hope this does not happen often. Oleg Alexandrov (talk) 04:20, 4 February 2008 (UTC)[reply]

Much respect, Oleg. I hope you are well. Well, I hope you reconsider this revert of set (or you that are just trying to illicit a better reformulation). I don't know how much simpler to state this: "A set is an abstract object." That also is the most fundamental account of what a set is. Are you concerned about simplicity or perhaps something else? The fact that a set is an abstract object is not a trivial observation. There a thousands of abstract objects that have an article, and yet I have only seen fit to make sure that theorem, and set specifically state it (i.e. it is important to me, because it is important to understanding the concept.) For some, this one clarification will be the key to understanding it. Please don't take the position that it just isn't important. It really is the first thing to say about what a "set" is. As usual, I am fascinated, and wondering about what this sort of move is all about.

I really would like to be able to clarify the metalogical, metamathematical, and yes philosophical foundations of many topics, however I get the sense that there is a general hostility to that whole project. Although, I think my formulation of theorem appears to be accepted (or tolerated?), for which I am grateful. Any help would be appreciated. Pontiff Greg Bard (talk) 17:36, 4 February 2008 (UTC)[reply]

Re. my revert. I think the first several sentences of an article should convey, in a language as simple as possible, the main ideas of the article. The main idea of a set is that it is a collection of objects. To start the article by saying that a set is an abstract object, is in my view, not so relevant, and rather confusing, to folks whose background may be less sophisticated than yours.

As such, while I understand your desire to "metalogical, metamathematical, and yes philosophical foundations of many topics", I would argue that the introduction is not the place for that. The introduction should be simple, even to the point of not saying things completely accurately. The goal is to teach the reader, not tell him upfront the most complete, accurate, and foundational issues. Any disagreements here? :) Oleg Alexandrov (talk) 03:35, 5 February 2008 (UTC)[reply]

Well, I agree on one thing: "I think the first several sentences of an article should convey, in a language as simple as possible, the main ideas of the article." So that leaves me mystified at the rest of your response.

First of all, the fact that a set is an abstract object certainly belongs in the lede, and most appropriately is the first thing to say about it at all. Abstract object is the primary catgeory of thing that this is. It is the first thing that was explained to me about the concept. It is the first thing I explain to other people when I explain to others about it. Perhaps your experience of this in mathematics does not include this, however the mathematicians do not have a monopoly on the academic anaysis of this subject. It is this last point especially, that I hope you and some of the other mathematicians will please consider.

I think you are looking at the word "abstract" and thinking that it is too abstract a word to use?! Please Oleg, when we are dealing with abstract objects it is very useful for our understanding to actually name what it is we are talking about so we can use our language to deal with it. I don't think this language is sophisticated at all (and especially as it compares to the mathematical jargon I see unmolested everywhere?!) It's just the language that is necessary to describe what it is. Wikipedia has wikilinks, so I think people can take it.

I also think that the "foundations" of a topic belong in the lede as well. The opening of every article should attempt to be a canonical account. I am not sure Wikipedia needs to "teach" as you put it, and WP:NOT tells us it's not a textbook. People should be able to learn on thier own. We should give them all the tools they need.

I just want people to understand. That is my primary motivation. Understanding right off the bat that what we are dealing with is an abstract object is very helpful. I think this helped to clarify the divisions over theorem (at least I hope it did.)

In an ideal world, all these articles would be reformulated by a team of Ordinary language philosophers at the end of the day.

I took a long time to get around to making that change. It seemed to me at the time that I didn't have a better formulation in mind myself, although I still wasn't entirely satisfied with the style. I figured that perhaps someone might work it in better than I did. I'm pretty sure there won't be any problem finding a source to cite on the issue if that is absolutely nessssary. At some point down the road, I'm going to put it back in there. If you have any ideas about it, let me know. Be well. Pontiff Greg Bard (talk) 05:39, 5 February 2008 (UTC)[reply]

Your intentions are nice, but the end result will be that you will turn off the reader right from the first sentence. [Heb 5:12-14]: “And ye are become such as have need of milk, and not of strong meat.". :)

Most people won't even know what the word "abstract" in the mathematical sense means. The concept of "set" is simpler to me than the concept of "abstract". Why use something complicated to explain something simple? Oleg Alexandrov (talk) 05:54, 5 February 2008 (UTC)[reply]

(Hopefully I am not butting in.) The edit appears fairly minor, and probably the two positions can be merged easily. I can agree that "a set is an abstract object" is one of the first things we learn about sets, since they are quite often once of the first "abstract" things you study in math (using mathematical language here). I think the part of the edit that makes the article less simple is linking to rather refined, though fundamental, philosophical articles in the lead of an article on a mathematical subject. Since set theory, especially naive set theory, has caused quite an examination of the philosophical foundations of mathematics, such links are appropriate *somewhere* in the article. I think the only concern is that it not be in what is more or less the opening sentence.

Could both agree to "In mathematics, a set is an abstract object that can be thought of as any collection of distinct objects considered as a whole." as the opening sentence (with only the single wikilink), and then an addition, perhaps to the Definition section (so quite fundamental to the article, and appearing quite early), links to the philosophical concepts being used somewhat informally there? I think the tone in that section is quite suitable for wiki-links to the philosophical articles that provide a sounder understanding of the terms. Alternatively one could follow a somewhat common style of having a two or three paragraph lead that has sentences for what it is, what it is used for, its history, its foundation, and its generalizations. I think a very short lead is much nicer, and that editing in the Definition section would probably reach consensus more quickly, but I don't see any insurmountable trouble with expanding the lead. JackSchmidt (talk) 06:08, 5 February 2008 (UTC)[reply]

Let's move this to Talk:Set#Set as an abstract object. Oleg Alexandrov (talk) 06:24, 5 February 2008 (UTC)[reply]

hi. it would be lovely if you could indeed try to retain the extra information that i put into the article by copy-paste instead of simply redirecting - thanks! -- Kku (talk) 09:34, 6 February 2008 (UTC)[reply]

Re my redirect. I have not seen much to merge. The external link was not that valuable from what I saw. The formula you put in has a bug too, at the denominator there should be ${\displaystyle \max(|x|,|y|)}$ instead of just ${\displaystyle \max(x,y)}$ I think, otherwise note that ${\displaystyle \max(-1000,-0.00001)=-0.00001}$ which I think is not what is intended there. You're welcome to add anything you find useful to the larger article though. Oleg Alexandrov (talk) 15:32, 6 February 2008 (UTC)[reply]

I have been using {{refimprove}} when I see multi-paragraph articles with only one or two references. If {{refimprove}} is not appropriate, what should I use? Do I just use {{unreferencedsection}} and {{citation needed}} tags. Also, can you point to a Wikipedia page that states whether what you stated on my talk page is just a guideline or a rule? 155.198.204.98 (talk) 10:32, 6 February 2008 (UTC)[reply]

What you did was fine. There is no firm rule about where to put {{refimprove}}. I feel that it is more appropriate at the bottom, where it is less distracting (such templates are for editors, not for readers). If course, if the article has no references at all or if it has dubious statements, then a note on top would work better. Oleg Alexandrov (talk) 15:34, 6 February 2008 (UTC)[reply]

I am going to be doing some more testing of the bot code on kiwix today. I've eliminated the query.php calls, changed the editing backend, and found another place where performance could be very easily improved.

I noticed there was often a long delay after bzip was called, and after some debugging I found out it was from the code that makes subpages for the logs. Testing shows that repeatedly catenating strings is very slow; see the code below. I think changing this code will have a useful effect in speeding up the bot run and reducing the overall CPU time. — Carl (CBM · talk) 17:49, 6 February 2008 (UTC)[reply]

Repeatedly catenating onto the end of a string is slow. This script takes about 14 minutes to construct the string $j, during which time it is CPU bound.

$a = 'x' x 100;

$last = time();
$first = $last;

for($i = 1; $i <= 100000; $i++) { 
  $j = $j . $a;
  if ($i % 1000 == 0) { 
    $new = time();
    print "$i " . ($new - $last) . " sec\n";
    $last = $new;
  }
}

print "Length: " . (length $j) . "\n";

$new = time();
print  "join: " . ($new - $last) . " sec\n";
print  ($new - $first);
print  " sec total\n";

Pushing onto an array and then joining the array is fast. This script takes under 1 second to run, and constructs the same string $j.

$a = 'x' x 100;

$last = time();
$first = $last;

for($i = 1; $i <= 100000; $i++) { 
  push @k, $a;
  if ($i % 1000 == 0) { 
    $new = time();
    print "$i " . ($new - $last) . " sec\n";
    $last = $new;
  }
}

$j = join "", @k;
print "Length: " . (length $j) . "\n";

$new = time();
print  "join: " . ($new - $last) . " sec\n";
print  ($new - $first);
print  " sec total\n";

Wow! Neat. Thanks for spotting this. Oleg Alexandrov (talk) 03:43, 7 February 2008 (UTC)[reply]

Hello! I think the bot you have set up is great for updating articles by quality, but at WikiProject Universities we want to use a different quality scale. We don't use the importance factor, and would also like to rate things like templates, categories, and featured lists. I updated our articles by quality statistics to reflect this change. Can your bot use this updated scheme?—Noetic Sage 00:59, 7 February 2008 (UTC)[reply]

Regretfully, due to the large number of projects (1176 at the moment), the bot does not support features specific to individual projects. If you'd like the the bot to support extra quality classes, you'd need to ask at WT:1.0/I and see if the folks there agree. At the moment there is no consensus on featured lists, and a proposal to add templates categories and templates failed in the past. But you can start a new discussion on that if you want. Oleg Alexandrov (talk) 03:48, 7 February 2008 (UTC)[reply]

Hello Oleg, I've noticed you around and would like to ask you to give me rollback permissions, so that I can more easily revert vandalism on math and stats-related pages. I understand any administrator can grant such privileges (see WP:Rollback). Thanks! --Zvika (talk) 13:13, 7 February 2008 (UTC)[reply]

Done. I would like to mention that it is very important that you do not use this feature except for reverting vandalism. It is very rude to be reverted by another editor with the rollback, any well-meaning (even if misguided) edit deserves an edit summary. Hope you'll use the feature for a good purpose! Oleg Alexandrov (talk) 15:47, 7 February 2008 (UTC)[reply]

Thanks. I will be sure to use it only in cases of obvious vandalism. Have a nice day, --Zvika (talk) 16:13, 7 February 2008 (UTC)[reply]

Thanks! I understand now. —Preceding unsigned comment added by 75.15.136.94 (talk) 05:47, 11 February 2008 (UTC)[reply]

As of recently, I noticed that something seems to be broken in WP 1.0 bot. Wikipedia:Version 1.0 Editorial Team/Portugal articles by quality statistics‎ is now always incomplete, missing the ratings as well as some categories like GA and FA. At 2008-02-04 it was okay, but after that it has been quite incorrect. Could you perhaps look into this issue?--Ivo Emanuel Gonçalves ^{talk / contribs} (join WP:PT) 16:55, 14 February 2008 (UTC)[reply]

The bot appears to have malfunctioned at Wikipedia:Version 1.0 Editorial Team/Biography articles by quality/173 as well. Ioeth _{(talk contribs friendly)} 18:02, 14 February 2008 (UTC)[reply]

Category:Top-importance Portugal articles and the other ones are empty, that's why the bot does not show them (perhaps there is some issue with the rating template your project uses)?

About the bio page, this fixed it. It was not a bot issue, it is just the Wikipedia server could not parse correctly the html with that banner. Oleg Alexandrov (talk) 04:11, 15 February 2008 (UTC)[reply]

I was going to say "what the hell?" after seeing that page empty, but then I noticed that that is not the category's name we are using. We add the "-related" to "Portugal", i.e. Category:Top-importance Portugal-related articles. Is this wrong? We certainly cannot have duplicate categories, especially when only one is working. I assume Oleg has far more experience than I do on this issue, so could you perhaps recommend a solution?--Ivo Emanuel Gonçalves ^{talk / contribs} (join WP:PT) 17:01, 15 February 2008 (UTC)[reply]

Well, the bot expects consistent naming. So, either you use "Portugal articles" everywhere, or otherwise "Portugal-related articles" everywhere. There's no way around I think. Oleg Alexandrov (talk) 05:29, 16 February 2008 (UTC)[reply]

We have always used "Portugal-related articles" since the project inception. If we could keep that without breaking the bot statics, it would be great.--Ivo Emanuel Gonçalves ^{talk / contribs} (join WP:PT) 15:15, 16 February 2008 (UTC)[reply]

Well, then look at Wikipedia:Version 1.0 Editorial Team/Portugal-related articles by quality statistics. :) Oleg Alexandrov (talk) 16:28, 16 February 2008 (UTC)[reply]

If I understand what appears to be the edit history, you are the author of the page at Image:Gaussian 2d.png. It looks as if the level sets are circles. That could lead the reader to think that the level sets will always be circles, whereas in fact any ellipse is a level set of a Gaussian function with a 2-dimensional domain. Examples should not be obvious exceptional special cases. If it's not too much work, can you tweak something in your software and put in an image that doesn't look circular? Michael Hardy (talk) 02:25, 20 February 2008 (UTC)[reply]

I did not create that image, I just moved to commons (you can see the original creator if you visit the image page). It is easy enough to make the gaussian elliptical (you could do that yourself even, the MATLAB code is on the image page, and U of M computers have matlab installed :) I'll try to get to it some time soon otherwise. Oleg Alexandrov (talk) 05:17, 20 February 2008 (UTC)[reply]

On User talk:Dino, you note,

Thank you for the new Image:Graph of sliding derivative line.gif. I have two requests. First, would it be possible to post the image source code on the image page, it is in the spirit of Wikipedia and may help other people understand how you made the image. Second, your blue curve is kind of wiggling, which it should not. Could you recreate the image so that it does not do that? Thanks a lot. You can reply here. Oleg Alexandrov (talk) 05:29, 20 February 2008 (UTC)[reply]

The "wiggling" bugged me too. I thought the image small enough it would be invisible, but I guess you've got good eyes. I tried many things, finally posting a question on SAGE forums. As to the source code, sure, I'll post it. But that will wait, as there's no point now. I'll see what answer I get.

dino (talk) 19:50, 20 February 2008 (UTC)[reply]

Well. You may never touch SAGE, but you improved it. I wrote a message to sage-support; someone took out a ticket, someone fixed it, and it will be in the next release of SAGE. Meanwhile, I used a work-around. Will post the generating code I used shortly; should be there by the time you read this. The page derivative should look nicer.

Thanks for nabbing this thing I tried to ignore.

dino (talk) 01:54, 21 February 2008 (UTC)[reply]

Cool, thank you! Two more things, if you don't mind. The image is very large, 842 KB, it would take a long time to load (even with broadband, the Wikipedia server is kind of slow). Can you use fewer frames in the figure, say 30-40, instead of 125. A smaller piece of the curve could be used too, and a lot of the whitespace could be cropped. Ideally having the gif under 250K would be great, I think. This will be some effort, but could make for a much improved experience of people seeing the animation I think. Oleg Alexandrov (talk) 04:47, 21 February 2008 (UTC)[reply]

Changed a few constants in the program & replaced image:Graph of sliding derivative line.gif with an smaller & faster image of about 208 kb.

dino (talk) 02:52, 22 February 2008 (UTC)[reply]

Wow! Thank you! Very nice addition to the page! Oleg Alexandrov (talk) 03:37, 22 February 2008 (UTC)[reply]

Hi, I'm part of the WikiProject Economics and I was wondering if your bot could split up our articles into categories such as "FA-class, Top-importance" articles rather than just FA-class, Top-importance, etc.? Meaning, I want to have categories for every intersection of class and importance; I've seen your bot do it for other projects, so it'd be great if it could be done for ours, too! Thanks! (Please post {{talkback}} on my Talk page when you've responded.) Gary King (talk) 16:00, 26 February 2008 (UTC)[reply]

I have created the categories and categorized the articles already now (Category:FA-Class, High-importance Economics articles, etc.) so all I need your bot to do is link the intersecting numbers in our project's chart (found at Wikipedia:Version 1.0 Editorial Team/Economics articles by quality statistics) to these categories. Gary King (talk) 16:33, 26 February 2008 (UTC)[reply]

Unfortunately the bot can't do category intersections, I don't recall the bot doing this for any projects. If you'd like this feature, you can request it at WT:1.0/I, to see what other people say. But I don't see any easy way to implement this, and am not sure how much people want it. Oleg Alexandrov (talk) 04:26, 27 February 2008 (UTC)[reply]

Huh, interesting. I've seen a table created by your bot (at least I think it's your bot) but I can't recall which Project I saw it on. I'll post back when I find it again. Anyways, I modified our Project's template so that it automagically creates the intersection category, so at least that part is already done. Gary King (talk) 04:29, 27 February 2008 (UTC)[reply]

Hey Oleg, I just wanted to inquire if you made a decision on whether or not the 1.0 Bot is going to place FL-Class in the stats table. I think at the project talk page, there was consensus to add it and just let the WikiProjects decide for themselves. I would really love to see this implemented, especially for WikiProjects like WP:NFL who have a lot of FL's. Thanks in advance. « Gonzo fan2007 ^{talk ♦ contribs} 00:40, 27 February 2008 (UTC)[reply]

Yes, I was planning to get to this today. Done now, see here. It will take a few days until the bot will start again, then it will add FL-class articles. Oleg Alexandrov (talk) 04:19, 27 February 2008 (UTC)[reply]

Awesome!! Thank you so much! « Gonzo fan2007 ^{talk ♦ contribs} 05:42, 27 February 2008 (UTC)[reply]

Hi Oleg,

Just wanted to stop by and let you know that an optional parameter has been added to {{Afd top}} to make it a bit easier to close. There is an ANI thread soliciting comments on it, but there should not be any difference in the resulting wikicode using the new parameter, so hopefully Mathbot won't be able to tell the difference and will happily go on doing its duty.

The thread on the template talk page can be found here. Thanks! --jonny-mt 15:45, 27 February 2008 (UTC)[reply]

I'll keep an eye on this. I think it won't cause trouble. Oleg Alexandrov (talk) 15:53, 27 February 2008 (UTC)[reply]

Oleg, I was wondering if it would be possible for your bot to keep count of Ohio articles at Category:Ohio for the new Portal:Ohio as it does for the math portal. If not, could you direct me to someone who might be able to help me?

Thanks for your time, Stepshep (I would sign with the four little ~s but my sig is kinda broke.)

I'll be very happy to give away the bot source code and help anybody set the bot up (the source code is linked to from User:Mathbot). However, I would not be able to run the bot myself, since in order to get a quality list the bot updates need to be monitored every day, and poor additions removed (so it is not the bot-writing part which is hardest, but monitoring it regularly).

The bot is written in Perl and should be able to run on any operating system. Do any Ohio-project folks know a bit about programming and would be willing to give it a try? Oleg Alexandrov (talk) 06:36, 28 February 2008 (UTC)[reply]

I'd be happy to try it, thanks for the link! I'll read a bit and see how to set it up.

Thanks §tepshep • ¡Talk to me! 14:53, 28 February 2008 (UTC)[reply]

Ya, that almost broke my brain. Is there a tutorial anywhere on how to run all of that? You must be one smart dude. §tepshep • ¡Talk to me! 22:29, 28 February 2008 (UTC)[reply]

There were Readme files included there. But you do need to have some Perl experience to be able to run the bot and taylor it to Ohio articles from math articles. It can be some work, but easier than writing your own bot from scratch. You can start by downloading the directories "modules" and "mathlists" from here, and try to run the "update_mathematics.pl" script. If you get that far, we can discuss what to do further. Oleg Alexandrov (talk) 03:52, 29 February 2008 (UTC)[reply]

Is it possible to adjust the WP 1.0 bot output for some of the templates?
The Latvia template has List-class articles included to the table output, whereas the Sweden and Philately templates does not. Philaweb T 13:45, 28 February 2008 (UTC)[reply]

The bot does not include a class if there are no articles in that class. Try to add some list-class article to Sweden articles, then hopefully the bot will reflect that in the table. Oleg Alexandrov (talk) 03:48, 29 February 2008 (UTC)[reply]

Hi Oleg,

You have created a nice picture of scattering at http://en.wikipedia.org/wiki/Image:Scattering_theory_illust.png

Would it be too much of a problem to add a similar plot of the scattered field, i.e., incident, total, and scattered fields separately? I have never used Matlab and am afraid that doing this would take me forever.

Please respond to (address removed).

Thank you! Misha Mishchenko —Preceding unsigned comment added by 169.154.204.2 (talk) 23:25, 28 February 2008 (UTC)[reply]

I plan at some point to replace that picture with an animation. Then I will also add a picture of the scattered field. I have little time recently for that, however, I promise I'll get to it in a week or two. (I also replied by email.) Oleg Alexandrov (talk) 03:46, 29 February 2008 (UTC)[reply]